Human Interface and the Management of Information. Methods, Techniques and Tools in Information Design

Frameworks and patterns on integrating usability activities into organizational work practices have been developed during the last years. However, usability and user-centered design activities remain fairly unknown to small-sized software development organizations. Empirical research on initiating usability activities in such organizational contexts is also limited to few cases. We present a case project in which our usability research team had six months to provide a small-sized software company with knowledge on how to improve usability of one of their products. Our approach is based on patterns of integration, on a selection of user-centered methods and on producing tailored usability artefacts.

This paper describes psychophysical experiments which were conducted to evaluate the influence of haptic information on visual perception of three-dimensional (3D) interpretation of an object’s shape in virtual environment. In particular, we investigated whether haptic information provided by a force-feedback device contributes to 3D interpretation of ambiguous visual patterns, which causes spontaneous alternations of bi-stable 3D percepts. The subjects’ task was to report by pressing a key which 3D shape was perceived when they were touching the virtual haptic cube along a pre-defined trajectory on the cubic surface. To examine haptic influence on the bi-stable percepts, the duration of each percept was recorded. The results indicate that haptic information of the surface shape can impose a dynamic constraint on visual computation of 3D shapes. The evaluation methods and results could be used for developing human-machine interfaces that provide more natural and realistic sensation of 3D objects in the future.

Aporia

is a Greek word meaning helplessness or difficulty in dealing with, or finding out about, something. This investigation is aimed at determining the aporia existing in the present maps with satellite and traditional pictures. Moreover, we present a quality metric called

antinomy

, to eradicate the aporia in the design of interactive systems with cartographic contents. The present quality metrics can be applied to the digital maps. The importance of traditional iconography is also stressed in the maps and the need to carry out heuristic assessments before using a new iconography. We need to increase the quality in hypermedia systems, communicability and efficiency of the digital maps, quite aside from location, the multimedia technological support used in the presentation of the cartographic information into computers, palms, mobile phones, etc., and potential users of the interactive systems.

An overview of how a model based specification approach can be used in the domain of automotive human machine interface (HMI) development is presented. The common paper based specification approach is compared to a model based, tool supported process. Requirements from different stakeholders for such an approach are outlined. Intended audiences are all stakeholders involved in the creation of graphical user interfaces ranging from design, usability engineering, and prototyping to specification and final product realization.

This paper reviews research in both information content and perceived informativeness in the literature, and examines the causal effect of two information content factors on perceived informativeness. A 2x2 factorial design was adopted in an experiment involving a hypothetical online retailer. Results from 120 surveys collected show strong support of the two hypotheses in the expected direction, i.e., both price and quality information had a significantly positive effect on perceived informativeness. Data also indicate that perceived informativeness is a significant predictor of visitor attitude toward the site and visitor intention to revisit.

Progress of ubiquitous computing oriented ICT promotes development of information appliances with various functions. In spite of the progress, most of user cannot receive the benefit of them. In this paper as the reason of it, the problems related to increasing enormous amount of information contents were discussed. Based on the problems, there are new challenges for designing user experiences; interactions with vast amount of information contents and integrations ubiquitous computing applications into daily life. Analyzing these challenges from view of human-centered design could derive the necessity to services design based on ubiquitous technologies and validation of them.

In a ubiquitous society, it is possible to use many kinds of information anytime, anywhere. Increasingly reliant on their mobile phones and PDAs, people often use information in public spaces. However, there are risks to entering highly confidential information such as personal data into a system. People have a strong awareness about the value placed on personal information, and worry about their information being leaked to people they don’t want to see it. In public spaces people also worry about real-world leakage (i.e. by non-digital means). In spite of the security provided by law and technology, people cannot handle information with reassurance. Complete security means both communication security and physical security. The purpose of this research is to construct an environment where ubiquitous services can be used with reassurance. In this paper we describe our first experiment in this research area.

The aim of this research is to suggest a most effective method for measuring user’s emotions expressed while using a product. This study developed a tool that effectively measures the user’s emotions expressed while using a product and that can complement the limitations of the psychological measuring method. The developed emotion logging software named VideoTAME, basically asks the subject to watch a recorded video clip of the user performing a specific task and to examine the emotional changes that had occurred. And a physiological measurement method that measures the user’s emotions expressed during product use with VideoTAME and that is easily accessible in the design field was suggested. By upgrading VideoTAME and overcoming its limitations as a psychological measurement method, and also by using the physiological measurement method mentioned this study to measure the user’s emotional changes, the correlation between the product usability and the user’s emotion will be able to be defined more clearly.

This paper was focused on the next generation of ubiquitous services by using ubiquitous networks and devices. This paper was especially focused on the transmission services of feelings to others. In the paper, some conventional transmission services of feelings were introduced and a few of the unconscious transmission services of feelings to both specific targets and multiple targets were pointed out. The paper proposed two examples of the new services that would enable the users to transmit their feelings unconsciously to others by using ubiquitous networks and devices. These services were proposed through the PRESPE (Participatory Requirements Elicitation using Scenarios and Photo Essays) approach and called respectively the “aura transmission system” and “back scratcher system”. Further discussion about the usefulness and possible negative influences on human nature or society by the new services was done, and the future research efforts of these services were described.

The purpose of this study was to test the hypothesis that nurse perceptions of technology they use in practice would affect their perception that they were able to provide high quality patient care. A survey assessing the variables was administered to 337 pediatric nurses from two academic freestanding pediatric hospitals in the US. Two separate equations were constructed, one to test whether technology perceptions affected individual quality of care and the other to test whether technology perceptions affected quality of care provided by the nursing unit. Nurse confidence in their ability to use hospital technology and their beliefs that the technologies were easy to use, useful, and fit their tasks are important predictors of nurse beliefs that they are able to provide quality care to their patients.

We are developing an intuitive information design that can increase the user’s feeling of reassurance and security when handling private information in public space. We conducted an experiment to collect user responses to displays providing awareness of the activity behind them while they handled private information. Two approaches were examined: a simple two-color LED array and a mirror. The former yielded less distraction and faster response speeds while the latter was more intuitive and offered more reassurance.

The purpose of this research is to establish a design methodology for collaborative systems (artifacts for collaboration), which utilizes one of the social analyses called Ethnomethodology. So far, the author proposed guidelines, principles, and a model as components of the design methodology based on examples of collaborative assembly of an everyday product and remote instruction, which uses a movable laser pointer. This paper proposes a simple way to adapt previous results: it is the design matrix which can be used in the requirement acquisition phase, the design phase and the evaluation phase. The matrix supported its usefulness in the collaborative system development.

It is well recognized that the evaluation of the expert systems is very important and difficult. This paper relates to the important issue of Expert System evaluation. A methodology for addressing the problem is presented. As evaluation model is proposed with the following main criteria: Usability; impact of the system in the behaviour and decisions of the user; Impact of the quality of the information provided to the user; financial impact; usefulness.

In this paper we detail a number of recommendations regarding how to evaluate interfaces to systems providing public access to environmental data and information. The primary focus is based upon web and Internet-based interfaces. These recommendations will assist data providers and information managers to evaluate the effectiveness of their current data access interfaces. The paper also outlines some guidelines on how to ensure these software interfaces are accessible to all citizens.

Whilst much effort is being invested into the design of human-computer interfaces that provide users with access to large amounts of information, there is less understanding of how human performance can be optimized in such interaction-driven contexts. We suggest one way to improve performance is to tailor the structure of the interface to the adaptive nature of cognition and orient behavior towards certain strategies, and away from others. Here, we show how small changes in Information Access Cost (IAC) - the time, physical and mental cost of accessing information - can effect powerful changes in cognitive processing strategies that subsequently affect performance. Increasing IAC on task-critical information was generally shown to encourage a more ‘memory-intensive’ processing strategy. This strategy not only improved future memory for task-relevant information, but also improved learning during problem solving, and planning. Implications for display designers in HCI are discussed.

An evaluation method of detection probability has been proposed by the authors for the measurement and evaluation of the industrial products based on the human performance by perception. As a practical application, the distributed parameters of psychometric curve were estimated by the experimental data concerning the three kinds of atribute of size, flatness degree and grayscale. As an example of experiments in the proposed methods, single madality and multiple dodalities sensory tests are applied to measure the perception performance of panels concerning the three atributes of FRP (fiber reinforced plastics) poroduct.

The limited screen space in small technical devices imposes considerable usability challenges. On the one hand objects displayed on small screens should be big enough to be hit successfully, but also small enough to house several objects on the screen at the same time. However, findings up to now show that single pointing is more effective in a large display compared to a smaller display. In the present experiment this was also confirmed for an applied multidirectional serial pointing task. Especially in more difficult tasks, results point at a shift of the speed-accuracy tradeoff. In large displays a fast and comparably accurate execution is chosen in contrast to a very inaccurate and time-consuming style in small displays. From an ergonomic point of view the outcomes recommend an optimized balance of task difficulty and display size in small screen devices.

Popular online services, such as Amazon.com, provide recommendations for users by using other users’ rating scores for items. In this study, we describe three types of rating systems: score-rated, count-rated, and digital-rated. We hypothesize that digital-rated systems provide the most useful recommendations. Then we analyze the differences in the results of the rating when the granularity of the score changes. Finally, we visualize users by developing a 2-D visualization system that uses a multi-dimensional scaling method.

Based on the ongoing development of a portal intended for use during the upcoming Olympics event in 2008, the portal’s main purpose is to allow volunteers, spectators, or any other participants of the Beijing Olympics to upload self-directed video clips and relevant advertisement clips associated with the event to the portal. Previous work has found dissimilar experiences in interacting with the portal – thus the need to use collaborative scenario building as part of the design process of the portal. The paper introduces the main purposes of the portal, and demonstrates how the intended outcomes of the portal and dialogues generated lead to the need for participatory design. The sessions building collaborative scenarios are also discussed, demonstrating how they can be used to guide stakeholder participation. Lessons learnt from the design exercise are discussed as a concluding note to the paper.

Recently, as the user-centered design has appeared to production design area as an important management philosophy, there has been various trials to reflect the user’s needs in production in many aspects. However there are various kinds of related inquiry methods and each method has different characters according to the data traits, the participant’s condition of inquiry, and the participant’s number in needing inquiry. So it is very important to choose appropriate method fits in the inquiry purpose and the proceeding situation so that these inquiry methods are applied to mobile phone domain. Thereafter this study addresses the simple cases of nine inquiry methods which were applied for the purpose of finding out mobile phone needs/ new function, and then describes the result of this research as the following subjects. What significant data does each method have contrasted with the others? What methods can make out synergy effect when they match together? What methods can be omitted and are necessary to extract variable data from participants according to the purpose of inquiry?

Designing mobile services is fundamentally different from designing online services. Not only are there differences in underlying technologies, but also in the way people use the services. If these differences are not taken into account, mobile services are likely to fail. However, it is quite difficult to provide a single-layered user interface in a small screen. This study aims at examining how design attributes of user interfaces of mobile services affect customer preferences. In order to explore customer preferences to each design attribute of mobile services, we measure and analyze customer’s WTP (Willingness To Pay) toward their different interface designs. Ultimately, throughout the study, we try to answer how to design mobile services in small screen of mobile devices. In addition, we propose an optimal design solution that customers likely prefer.

In human-human communications, especially in face-to-face communication, sub-verbal and non-verbal messages have more importance than messages transported by words. On the other hand, in traditional man-machine interfaces, machines only understand pre-defined operations, and never understand operators’ sub-verbal and non-verbal messages. This causes some usability problems in machine operations. We have proposed elsewhere that Grasping Information Interface Concept (GIIC) is the key idea to alleviate the above-mentioned communication gap between man and machine, and this paper verifies GIIC experimentally. GIIC regards grasping-and-moving as a fundamental hand operation necessary for performing tasks using modern man-machine interfaces, and behavioral measures associated with grasping, such as force, posture, etc., should have much potential in developing task-adaptive and operator-adaptive interfaces because it is known that how people grasp devices is dependent on the purpose of these tasks. GIIC provides machines with a communication channel to understand operators’ intents more through the grasping information.

The accessibility of information on web-space is considered to be focused on the properties of hyperlinks. Although the reasons to put in a hyperlink are multifarious, only one kind of link exists in HTML structure. As a result of this, users have to search for links by using syntactically techniques so as to retrieve information in web-space. In this paper, four kinds of hyper links are introduced into HTML, in order to improve the accessibility of information in web-space: External links, Internal links, Navigation links and Intra-unit links. Designers and users can share the image of the link target page by using these kinds of links.

This paper aims to investigate collaborative product development activities and highlights the importance of human perspectives for developing a collaborative product development (CPD) system. For CPD systems to be useful, it is necessary to effect usable and useful interaction. Therefore, the human-centred design (HCD) method had been implemented into the developing process of a CPD system. The modified HCD method specifies how the activities fit into the overall system development process. A proposed plan guides: (1) the design process activities described, (2) procedures for integrating these activities with other system development activities, (3) the individuals and the organisation(s) responsible for the design activities and the range of skills and viewpoints they provide and (4) effective procedures for establishing feedback and communication on design activities as they affect other design activities and methods for documenting these activities.

This paper describes the basis content of Artificial Psychology as well as Affective Computing and Kansei Engineering and other technology that are correlated with the research. Also it reviews the research history of Artificial Psychology theory and practice, and the future research direction of the Artificial Psychology is indicated.

This paper describes a HCI system for the control of home appliance which is focused on human friendliness. This system utilizes two USB cameras to enable a user to select home appliance easily by hand pointing gesture. We propose two different methods of storing the three dimensional position of home appliance by user. Home appliance selection method and feedback for user’s wrong pointing direction are also described in this paper. Because of the low cost installation, simple operation and interactive feedback, our proposed system enhances the usability and human-friendliness.

In this paper, we present that Enterprise information systems (EIS) can be abstracted to a scenario in which users could schedule the task suites of controlled entities under enterprise security mechanism, and propose a novel Entity Driven Task Software Framework (EDTSF) for EIS. Based on existing hierarchical information system architectures, the EDTSF could clearly implements enterprise business partition, reduces coupling between data objects and tasks; it also increases flexibility and expansibility of the EIS by expanding service function of data object and reusing the scheduling mechanism at system level. We have noticed that little research has been done on the partition rules and methods of enterprise business. The EDTSF is a new software framework and its application results show that the EDTSF is an effective approach to analyzing, designing and implementing EIS.

Technology roadmaps are often referred for better decision making by HCI professionals who connect human factors with product development and innovation. We conducted user study that explore information seeking and tracing behaviors in using technology roadmap. The research revealed that HCI professionals exhibit distinctive patterns in using technology roadmap, depending on their expertise in technical knowledge and work experience. Finally, we designed new user interface of an interactive technology roadmap system based on the research findings. We demonstrated its usefulness in seeking task-dependent information, intuitiveness in information visualization, and easiness to use as a reference for technologies.

A large number of the world’s cultural heritage sites and landscapes have been lost over time due to the progress of urbanization. Digital archive projects that digitize these landscapes as virtual 3D worlds have become more popular. Although a large numbers of studies have been made on reconstructing 3D virtual worlds, the previous methods have been insufficient, because they require significant effort. In this study, we propose a new method of reconstructing a 3D virtual world only from photo images that requires little intervention. The idea is to reduce the learning curve of the software need and automate the method as much as possible so that we can digitize as many heritage sites as needed. In our approach, we first reconstruct 3D models from single 2D photos using an image based modeling and rendering(IBMR) technique. After reconstructing models from all the available photos, we connect the 3D models into one unified 3D virtual world. Specifically, we implemented a seamless connection algorithm that supports free viewpoint translation. And we demonstrated the reconstruction of part of a cultural heritage site based on our system.

Knowledge Discovery in Databases (KDD) is an active research domain. Due to the number of large databases, various data mining methods were developed. Those tools can generate a large amount of knowledge that needs more advanced tools to be explored. We focus on association rules mining such as “If Antecedent then Conclusion” and more particularly on rules visualization during the post processing stage in order to help expert’s analysis. An association rule is mainly calculated depending on two user-specified metrics:

support

and

confidence

. All current representations present a common limitation which is effective on small data quantities. We introduced a new interactive approach which combines both a global representation (2D matrix) and a detailed representation (Fisheyes view) in order to display large sets of association rules.

Development of intuitive visualizations requires a systematic approach that includes a focus on the user. Creating interactive visualizations for complex systems often requires the integration of information from existing systems and sensor data to provide the operator with real-time information. The objective of this research was to fuse information from sensor technology with flightline maintenance information to support aircraft maintenance logistics. The research was conducted in two phases. A user-centered approach was used to design visualizations in each phase; however, in Phase II a cluster analysis technique was utilized to support the design. User feedback indicated that incorporating a technique to map data and decisions resulted in interactive visualizations that were well accepted by users and provided the important information needed for their decision making tasks.

Today’s desktop computers can be regarded as evolution of the typewriter. They are well suited for office applications but far from optimal for e. g. mechanical design, traditionally conducted on large drawing tables. Large screens are nowadays widely available for table-like computer workplaces. But their resolution is still too poor. To overcome this drawback the »fovea-approach« has been developed. A slim tablet PC (the so called Fovea-Tablett

®

) can be put deliberately on top of a large table screen. The position of the tablet in relation to the table screen is tracked and the screen content of the table display just below the tablet is displayed on the Fovea-Tablett

®

: just as if one would look through the tablet onto the table but with higher resolution. Whereas the table display is still good for overview, the Fovea-Tablett

®

brings highest resolution to the region, one is just focusing on.

In this paper we describe a first version of ZEUS, a web application that combines browsing, searching and object presentation. With the zooming and panning based navigation concept of ZEUS and a hierarchical organization of the information space we try to solve the problems of information overload. It has to be evaluated if categorization, zooming and a full text search can minimize that the user gets lost in hyperspace. The concept of ZEUS is based on some thesis about human cognition, navigation and exploration which we hope to prove with evaluation and user testing of our application in the future.

The aim of this research is to suggest folksonomy-based collaborative tagging system for supporting designers in group who interpret visualized information such as images through grouping, labeling and classifying for design inspiration. We performed field observation and preliminary studies to examine how designers interpret visualized information in group work. We found that traditional classification methods have some problems like lack of surface and time consuming. Based on this research, we developed PC based group work application, named I-VIDI. By implementing I-VIDI based on functional requirements, we have showed how I-VIDI reduces problems found from current image classification methods such as KJ clustering and MDS. In future case study, we plan to conduct extensive user research to evaluate the system further as well as adding more functions which can be usefully applied to collaborative design work.

The development of an in-store sales support system that focuses on the “virtual try-on” of clothing is the aim of the research project “IntExMa”. Based on sophisticated virtual reality technology, the interactive system provides visualization of made-to-measure shirts in combination with digital customer counterparts. The system is intended for seamless integration into existing processes at the point-of-sale and for the support of the collaborative consultation process between salesperson and customer. This paper describes the various system parts stemming from different research disciplines and their integration under the goal of high usability in an everyday setting.

Keeping sight of the enterprise’s workforce strengthens the entire business by helping to avoid poor decision-making and lowering the risk of failure in problem-solving. It is critical for large-scale, global enterprises to have capabilities to quickly identify subject matter experts (SMEs) to staff teams or to resolve domain-specific problems. This requires timely understanding of the kinds of experience and expertise of the people in the firm for any given set of skills. Fortunately, a large portion of the information that is needed to identify SMEs and knowledge communities is embedded in many structured and unstructured data sources. Mining and understanding this information requires non-linear processes to interact with automated tools; along with visualizations of different interrelated data to enable exploration and discovery. This paper describes a visualization solution coupled with an interactive information analytics technique to facilitate the discovery and identification of workforce experience and knowledge community capacity.

Scientists in the data mining field are constantly faced with the challenge of finding useful information from a huge amount of information. We have to analyze the data until we can get the appropriate information. We have to select one part of the data, compare them against each other, or arrange them in certain order. This approach is also known the trial and error approach. A trial and error approach requires the users’ judgment, for example, to correctly set certain parameters; it is an approach that place importance not only to the end result, but also to the process in achieving the end result. In this paper, we propose visualization methods to visualize past working history for supporting trial and error approach in data mining. We use our methods to visualize web browsing logs and data browsing logs in genome science fields.

Despite the burgeoning interest shown in visualizations by diverse disciplines, there yet remains the unresolved question concerning comprehension. Is the concept that is being communicated through the visual easily grasped and clearly interpreted? Given the vast variety of users and their visualization goals, it is difficult for one to decide on the effectiveness of different visualization tools/ techniques in a context independent fashion. To capture the true gains of visualizations, we need a systematic framework that can effectively tell us about actual quantifiable benefits of these visual representations to the intended audience. In this paper, we present our research methodology to establish a metrics-based framework for comprehension measurement in the domain of software visualization systems. We also propose an innovative way of evaluating a visualization technique by encapsulating it in a visualization pattern where it is seen as a solution to the visualization problem in a specific context.

TreeMap is one a common space-filling visualization technique to display large hierarchies in a limited display space. In TreeMaps, highlighting techniques are widely used to depict search results from visual queries. To improve visualizing the queries results in the TreeMap, we designed a continuous animated multi-distortion algorithm based on fisheye and continuous zooming techniques. To evaluate the effectiveness of the new algorithm, we conducted an experiment to compare the distortion technique to the traditional highlighting methods used in TreeMaps. The results suggest that the multi-distortion technique is only effective with small result sets but not as effective as simple highlighting for large search result sets.

Scroll bar is one of the most frequently used components of graphical user interfaces (GUIs). It is generally considered to provide overview + detail functionality. The scale ratio of a scroll bar refers to the rate of the document dimension to the proportion being displayed. It is usually determined by the length of the document and the height of the display window if considering the case of a vertical scrollbar. The user has no direct control to the scale ratio through the scroll bar, which is inconvenient when dealing with long documents in overview and navigation tasks. This inconvenience is more prominent for small display devices. This paper presents a novel GUI component, ActiveScrollbar, that enhances the standard scroll bar by providing direct scale ratio control without consuming more screen space. This component was expected to be more useful in hand held devices.

This paper proposes a system that automatically deforms the images transmitted in both directions so as to achieve two goals: easing the concerns of users and ensuring well enough support. We examine here the effect of visual information quality on the effectiveness and impression of collaboration between a novice user and an operator.

Poor data quality (DQ) has been a troubling issue within many organizations. Thus it is important for a decision-maker to be aware of the data quality of the information based on which he/she makes a decision. However, the mental integration of DQ metadata in to a decision process poses a significant cognitive challenge. A decision-maker could be overloaded and may not effectively take advantage of the DQ metadata [2]. This paper thus proposes a visualization approach to reduce the amount of cognitive resource required for such mental integration.

Human-Computer Interaction (HCI) is all about the way in which people interact with computer systems. This paper focuses on the cognitive aspects of HCI when a user is searching for information, so as to facilitate effective user interactions with vast amounts of available information. Search engines provide a ‘one-size-fits-all’ model, which do not adequately cater to the differing needs of searchers at different points in time (continuously changing situations in time/space, as per sense-making theory). We posit that from a system designer’s point of view, capturing the ‘user ability to specify his information need’ will help operationalize task/searcher characteristics (hence the user need) and help the designer provide better interfaces for search that fit the needs of the user and lead to search efficacy and searcher satisfaction. The study should advance HCI for search through greater understanding of user needs, enhance search interfaces and lead to theory development.

We propose an image gathering and display method for efficient browsing of surveillance camera images. The proposed method requires large cost to inspect lengthy image sequences taken by a surveillance camera. The proposed method involves generating a still image by gathering the moving parts from image sequences captured by a fixed camera. The gathered images are generated for several intervals of time and are displayed hierarchically. A user can easily browse the scene by observing the images with moving parts. Since detection and recognition of the target objects are performed by a human operator, efficient and reliable browsing can be established.

The previous study [9], [10] showed the fuzzy Bayes model successfully predicted print defects with a 50% hit rate at the first top prediction and an 80% hit rate within the top five predictions. However, the previous study was limited to English. In this study, Korean and English descriptions in predicting print defects by Korean subjects were evaluated based on fuzzy Bayes models. For the study, Korean descriptions were collected in Korea, and Bayes models were developed and evaluated. The result shows that Korean subjects much more accurately predicted print defects when they used Korean descriptions than English descriptions. Afterwards, English descriptions by US subjects will be collected, and both Korean and English lexicon data will be compared. Finally, the study will investigate a Korean-English cross language diagnosis (Q-KE-CLD) system to identify print defects based on the fuzzy Bayes model.

Search engines and metasearch engines on the World Wide Web typically display representative information such as the result document titles and snippets in a one-dimensional ranked list. Alternatively, a few others such as Clusty and Kartoo cluster their results so that users can, to a certain extent, interact with documents, keywords and/or clusters. The number of documents that can be effectively presented in one screen is usually limited to between 10 and 20. In this paper, we propose a method based on treemap visualization that substantially improve the information compactness. In addition, it provides a few unique post-search interaction methods that enable a user to manage a large number of results. An online prototype system is built. Various experiments are designed and done to evaluate the efficiency, effectiveness and usability.

This paper demonstrates a successful application of a Fuzzy Bayes machine-learning tool for classifying large amounts of narrative text, involving the use of ROC curves to identify optimum prediction threshold values at which to filter predictions for manual review. Different thresholds were used for different categories to optimize results and effectively minimize resources necessary for manual coding of a large number of claims narratives randomly extracted from a large U.S. insurer. The results indicated that utilizing a computer approach with strategic assignment of manual narratives filtered out approximately 15% of the narratives for manual review and resulted in a final accuracy at the two-digit classification level of 81%.

We discuss the interface of VISTO, a Content Based Image Retrieval system for vector images in SVG (Scalable Vector Graphics) format. The system includes different engines and one common graphical interface. In fact, to support the different requirements of different application domains, the system offers a variety of engines. Most notably, due to its modular architecture, the system allows users to add engines at runtime; the interface provides support for newly added engines, including parameter tuning. The interface is designed for two classes of users: application domain users and researchers in the field of multimedia. Application domain users can use both query-by-sketch and query-by-example to search collections; researcher users can test, tune, and compare engines, and they can design datasets to be used in batch mode. VISTO is an open source project developed in Java; it uses advanced features of the language, such as the Core Reflection API, to dynamically adapt the interface to the available engines. The system is being validated in different application domains, including the production of 2D animation, the Sign Writing Language, and the BLISS Language.

Current systems force users to store their documents in hierarchic filesystems. However, users remember their documents not in terms of ad-hoc categories, but of their contents, why they were written, etc. Describing documents using such autobiographic information would be a simpler and more effective way for users to retrieve them. Furthermore, we found that users mention printed versions of documents when retrieving them. However, current systems don’t regard the printed and electronic versions of a document as facets of the same entity. To solve these problems we developed an interface that allows users to browse their autobiographic information, gathered by a special-purpose monitoring subsystem, to retrieve their documents. We bridged the gap between the real and electronic worlds by maintaining the association between paper and electronic documents resorting to RFID technology. Tests show our interface allows users to efficiently retrieve their documents.

This paper proposes a new video digest and delivery system, “ChocoParaTV”, for applications such as blogs that include video content, and CGM sites. The video digests are generated based on automatic extraction of emphasized speech portions. To deliver the digest information and the digest to audiences, we also propose a new video digest description: CH-RSS; it can generate digested contents dynamically to match the user’s requested digest time. With the ChocoParaTV browser, the users first view the digest and then visit the original web site if they want to watch the original content. The effectiveness of the proposed system was confirmed by a field experiment; about 40% of the users accessed the original web sites through the ChocoParaTV browser digests. This indicates that our method is very effective in making video content attractive to the audience.

Semantic Text Portion (STP) is a text portion in the original page which is semantically related to the anchor pointing to the target page. STPs may include the facts and the people’s opinions about the target pages. STPs can be used for various upper-level applications such as automatic summarization and document categorization. In this paper, we concentrate on extracting STPs. We conduct a survey of STP to see the positions of STPs in original pages and find out HTML tags which can divide STPs from the other text portions in original pages. We then develop a method for extracting STPs based on the result of the survey. The experimental results show that our method achieves high performance.

Ontology is widely used in the areas of knowledge engineering, web based data mining, etc. In rough set theory, accuracy of approximation of set and a concept of granularity are introduced. Rough set theory is very useful to define dependency among attributes and extract decision rules from the set. One of our main aims of this paper is to propose a concept of rough ontology. A concept of rough ontology is extended concept of rough set, and it enables us to use flexible information system in the form of ontology. And rough ontology is useful to introduce concepts rough set theory in to ontology. In this paper we formulate a concept of rough ontology, which is extended concept of rough set theory. We define upper and lower approximation, accuracy of approximation of preference, concept of granularity of preference. And we also show the property of rough ontology.

For an advanced next-generation Human-Computer Interaction (HCI) interface, combining natural language processing (NLP) is an inevitable choice as human language is the most common and sophisticated communication device. Among various topics in NLP, word sense related topic is one of the most challenging areas. In this paper we present a method adopted from ACOM that automatically generate bilingual lexicon using aligned parallel corpus. The results of the test on the predefined test set for the English and French Bibles show the method correctly produce target words with 70% correct ratio. Besides, the proposed method generates target words that reflect contextual relationship between source and target words such as

garrison

and

Philistins

.

This study proposes a keyword search technique to reduce the traffic caused when applying Distributed Hash Table(DHT) base P2P systems as middleware for ubiquitous computing application in order to ensure effective search in Mobile Ad-hoc Network (MANET). For query with many keywords, nodes should be visited as much as the number of keywords and many posting files(inverted list) should be sent to the next node and these posting files bring about traffic. To solve this problem firstly the size of inverted list nothing to do with search result was diminished using distance of the hash values obtained by hashing keywords within document; secondly the positive false rate of bloom filter can be lowered since the size of posting file to be hashed can be reduced through distance-based bloom filter technique and by doing this the size of inverted list to be sent can be reduced as well. This technique is based on Plaxton’s approach and the proposed technique was evaluated through a series of tests.

We explore the idea of integrated physically based manipulation and decision-making tree for navigation to support design rationale. It introduces a way to support user participation. This paper also proposes and validates a framework which combines several techniques to aid user to make decisions. We hope this research can assist untrained users to quickly feel comfortable and satisfied for a computer aid system by rational way. This paper focuses on a system prototype for the client customization process in apartment plan design to exemplify our concepts.

This study uncovered critical domains and themes of compliments and complaints that influence consumers overall satisfaction with using virtual travel agencies. Four domains, namely, “customer Service & Support”, “trip schedule change”, “product experience” and “firm credibility” were identified as areas where problems arose and extreme dissatisfaction resulted. Three domains emerged as the best predictors for satisfaction: “product experience”, Customer Service & Sopecially at managerial level is critical for these virtual travel firms. The outcomes of this research lend insights into how to effectively manage consumer online reviews and turn this domain of valuable resources into knowledge and a strategy tool for the virtual travel segment of the hospitality and tourism industry.

This paper summarizes improvements to an earlier developed Fuzzy Bayes approach for assigning coding categories to injury narratives randomly extracted from a large U.S. insurer. Improvements to the model included: adding sequenced words as predictors and removing common subsets prior to calculation of word strengths. Removing subsets and adding word sequences improved prediction strengths for sequences found frequently in the training dataset, and resulted in more intuitive predictions and increased prediction strengths. Improved accuracy was found for several categories that had proved difficult to code in the past. This study also examined the effectiveness of a two-tiered approach, in which narratives were first categorized at the broad level (such as [falls]), before classification at a more refined level (such as [falls from heights].) The overall sensitivity following a two-tiered approach was 79% for predicting classifications at the broad category level and 66% for the more refined prediction categories.

Retrieval systems have become one of the most used categories of computer tools. Yet the interfaces of modern information retrieval systems fail to address the correlation between the user’s context and her information need. Furthermore, they usually do not integrate methods that allow whole communities or groups of users to profit of single retrieval instances. In this paper, we present our vision of an innovative, collaborative information retrieval and presentation approach based on the human factors balance model. Our approach combines automatic natural language processing results with handcrafted knowledge models and integrates implicit retrieval based on intelligent document segmentation and presentation, providing users with contextually relevant information.

This study compared the accuracy of three Singular Value Decomposition (SVD) based models developed for classifying injury narratives. Two SVD-Bayesian models and one SVD-Regression model were developed to classify bodies of free text. Injury narratives and corresponding E-codes assigned by human experts from the 1997 and 1998 US National Health Interview Survey (NHIS) were used on all three models. Using the E-code categories assigned by experts as the basis for comparison all methods were compared. Further experiments showed that the performance of the equidistant Bayes model and regression model improved as more SVD vectors were used for the input. The regression model was compared to a fuzzy Bayes model. It was concluded that all three models are capable of learning from human experts to accurately categorize cause-of-injury codes from injury narratives, with the regression-based model being the strongest, while all were dominated by multiple-word fuzzy Bayes model.

When a person requests, for example, “I want to see a bright and exciting movie,” the words “bright” and “exciting” are called

Kansei

keywords. With a retrieval system to retrieve recommended movies using these

Kansei

keywords, a viewer will be able to select movies that fit the

Kansei

without actually having to view samples or previews of the movies. The purpose of this research is to clarify a method to construct a support system capable of selecting movies that fit the viewer’s

Kansei

, and to verify the effectiveness of this method based on

Kansei

engineering, for the selection of recommended movies. To accomplish this, we extract the features of a movie using factor factoranalysis from data from a Semantic Differential Gauge questionnaire, then link the viewer’s

Kansei

with the features using multiple linear regression analysis. After constructing a prototype · system to verify the effectiveness, ten examinees viewed a movie selected by the prototype · system. “The selected movie fit the Kansei” at a level of about 70 percent.

List is a simple and useful graphical user interface (GUI) component for organizing linearly structured information when it contains only a few elements. However long lists are difficult to use because only a small part of the list is shown each time, to gain an overview or select elements from a long list the user usually needs to scroll the list many times. This problem becomes more serious as the list gets longer. This paper presents a novel solution, LensList, by applying the focus and context technique to the view of a list. LensList dynamically changes the elements to bigger sizes to form a focal area around the mouse cursor while keeping the elements in the peripheral area in smaller sizes as context. This enables it to display a longer list within the same screen area. Therefore it can be more efficient for performing browsing and navigation tasks.

It is efficient that free text interface is implemented in medical record when Health Care Workers (HCWs) want to write patient detail information. However contents of free text are not be able to evaluate easily. If we evaluate free text, we can use various natural language technique. Especially, knowledge of human recognition to free text is very useful, and we can construct a model of writing medical document. Concretely, this knowledge is “Context based Structure (CBS)” based on Diagnostic-Therapeutic cycle. On this time, we analyzed actual medical document (Incident / Accident Report in hospital) using this CBS and we made navigate system that was able to display something HCWs should write about according to judge contents.

Video retrieval continues to be one of the most exciting research areas in the field of multimedia technology. With the advancement of sensing and tracking technologies it is possible to generate multiple personal video streams during events from different perspectives. This paper presents the first prototype of the MyView system that provides on-demand personalized video streams with multiple cameras and additional sensors. This system captures and stores video streams in an indoor office, extracts high-level events from local positioning system (LPS) tracking information, and provides on-demand video segments for high-level queries by mixing the multi-camera video streams for a stream of the best view.

To reduce the workload of the driver due to the increasing amount of information and functions, intelligent agents represent a promising possibility to filter the immense data sets. The intentions of the driver can be analyzed and tasks can be accomplished autonomously, i.e. without interference of the user. In this contribution, different adaptive agents for the vehicle are realized: For example, the fuel agent determines its decisions by Bayesian Networks and rule-based interpretation of context influences and knowledge. The measured variables which affect the driver, the system, and the environment are analyzed. In the context of a user study the relevance of individual measured variables was evaluated. On this data basis, the agents were developed and the corresponding networks were trained. During the evaluation of the effectiveness of the agents it shows that the implemented system reduces the number of necessary interaction steps and can relieve the driver. The evaluation shows that the intentions are interpreted to a high degree correctly.

The Fraunhofer-Gesellschaft has selected “human-machine interaction” as one of twelve areas of technology with a particular potential for innovation and market-relevance. The paper gives a brief overview of the goals and research topics of the initiative.

We present a method for moving object contours detection based on spatial-temporal characteristics. Using S-T features, the contour of moving object can be well distinguished from background; therefore the moving objects are detected without the need of establishing and updating background models. The detection method can handle situations where the background of the scene suffers from the noises due to the various facts, including the weather condition such as snow or fog and flicker of leafs on trees, and bushes. The algorithm estimates the probability of observing pixel as a contour pixel based on a sample of intensity values for each pixel during a period of time and its local gradient in current frame. The experiments show that this method is sensitive to changes caused by moving objects and is able to avoid the affection of complex background. The paper also shows how to separate multi-person based on the contour detection results using template matching. The approach runs in real-time and achieves sensitive detection.

More Recently, researchers have explored to provide a queue management scheme with differentiated loss guarantees for the future Internet. Various types of real time and non-real time traffic with varying requirements are transmitted over the Internet. The sides of a packet drop rate, an each class to differential drop probability on achieving a low delay and high traffic intensity. Improved a queue management scheme to be enhanced to offer a drop probability is desired necessarily. This paper considers multiple random early detection with differential drop probability which is a slightly modified version of the MQDDP model, to get the performance of the best suited, we analyzes its main control parameters (max

th

, min

th

, max

p

) for achieving the proportional loss differentiation (PLD) model, and gives their setting guidance from the analytic approach. we propose Dynamic-multiple queue management scheme based on differential drop probability, called Dynamic-MQDDP, is proposed to overcome MQDDP’s shortcoming as well as supports static max

p

parameter setting values for relative and each class proportional loss differentiation. MQDDP is static according to the situation of the network traffic, Network environment is very dynamic situation. Therefore max

p

parameter values needs to modify too to the constantly and dynamic. The verification of the guidance is shown with figuring out loss probability using a proposed algorithm under dynamic offered load and is also selection problem of optimal values of parameters for high traffic intensity and show that Dynamic-MQDDP has the better performance in terms of packet drop rate. We also demonstrated using an ns-2 network simulation.

MPEG-21’s digital item adaptation technology becomes a new way for universal multimedia access. It needs transcoder to change media resource’s format and so on according to delivery context. Then, the use of heavy transcoder with various transcoding functions integrated into one altogether is so complicated and difficult in supporting universal multimedia access. Unit transcoder is useful is to resolve this question, in which a transcoder has only one transcoding function. This requires considering how to compose a set of unit transcoders. Thus, given a set for end-to-end different service quality pairs according to the character of application as defined by user, this study suggests how to compose complete unit transcoders that can always create one and more transcoding path(s) for each pair in the set. This method has a question of creating too many transcoding paths for each pair of end-to-end different service quality. Thus, this study also suggests the algorithm that generates minimum unit transcoder sets to support multimedia adaptation with minimum unit transcoder. The algorithm suggested was implemented into multimedia stream engine, and this paper describes the results of experiment for this algorithm.

This paper captures our experience developing Algorithms to solve Combinatorial Problems using different techniques. Because it is a Software Engineering problem, then to find better ways of developing algorithms, solvers and metaheuristics is our interest too. Here, we fixed some concepts from Knowledge Management and Software Engineering applied in our work.

Forms are ubiquitous in today’s software applications, so automation of form generation is highly desirable. In this paper, we provide improvements on Xforms model including data and event model. At the same time, we give a new method using use case for automatic form-building and the transformations from use case models to form user interfaces.

As computers are becoming more and more ubiquitous, moving from the desktop into the infrastructure of our everyday life, they begin to influence the way we interact with this environment - the (physical) entities that we operate upon in order to achieve our daily goals. The most important aspect of future human-computer interaction therefore is the way, computers support us in efficiently managing our personal environment. This paper addresses the fundamental components that are involved in the forthcoming human-computer-environment interaction.

Nowadays, one very complicated problem bothering network analysts too much is the redundant data generated by IDS. The objective of our system SATA (Security Alert & Threat Analysis) is trying to solve this problem. Several novel methods using data mining technologies to reconstruct attack scenarios were proposed to predict the next stage of attacks according to the recognition the attackers’ high level strategies. The main idea of this paper is to propose a novel idea of mining “complicated” attack scenarios based on multi-agent systems without the limitation of necessity of clear attack specifications and precise rule definitions. We propose SAMP and CAST to mine frequent attack behavior sequences and construct attack scenarios. We perform a series of experiments to validate our method on practical attack network environments of CERNET. The results of experiments show that our approach is valid in multi-agent attack scenario construction and correlation analysis.

In Korea, many small and medium size companies have introduced information system for their effective management. The Korean government have supported them to set this system. But many of these companies failed to utilize this system. One reason is high maintenance and customization cost. In this research, we developed an effective methodology for constructing an information system by themselves. And they can easily renew their system according to the change of their business process. This methology cannot support the internet version, but low cost and easy to maintenance will be the strongest point to set up an early version of information system for small size companies.

An interactive system in which the user can program animated agents visually is introduced: the Visual Agent Programming (VAP) software provides a GUI to program life-like agents. VAP is superior to currently available systems to program such agents in that it has a richer set of features including automatic compilation, generation, commenting and formatting of code, amongst others. Moreover, a rich error feedback system not only helps the expert programmer, but makes the system particularly accessible to the novice user. The VAP software package is available freely online.

MA(Mobile Agent) is active, autonomous, and self-replicable software object containing both computational logic and state information. One advantage of using MA paradigm instead of conventional message passing paradigm is that it can reduce communication traffic among the computing devices in the system. Also, the MA paradigm supports asynchronous interaction of computing devices, enabling more efficient system services. In order to adopt the MA paradigm into the ubiquitous computing technologies, it is necessary to develop a lightweight middleware platform, called MAP(Mobile Agent Platform), that supports agent migration and install it in various devices such as PDAs, hand-held devices, and digital appliances. In this paper, we propose our lightweight MAP, named

KAgentPlatform

, that is developed based on J2ME for ubiquitous environments. Especially, we describe our design and implementation of the

KAgentPlatform

, and show the experiments of adaptive UI migration based on ubiquitous mobile agents.

This paper describes about construction of Web application for cusp surface analysis. Client accesses to the Web server in order to analyze the data for cusp surface analysis, and a user’s browser downloads Web page for data input automatically. Client can be offered through Web browsers by using HTML generated dynamically by Java Server Pages (JSP) technology, or Java applets. Tomcat is operated as a plug-in of Apache to start the Java servlet on the server. Tomcat is the servlet container that is used to carry out cusp for Java. Cusp for Java is applied as Java servlet. It is a java class that extends a J2EE-compatible Web server. Cusp for Java receives HTTP (Hypertext Transfer Protocol) request from the browser, and provides views of result data as HTML.

In ubiquitous computing, most people need to track various sources of news using various devices, but it becomes difficult once there are more than a handful of sources. This is the reason for this is that users have to navigate to each page, load it, remember how it’s formatted, and find where they last left off in the list. To solve these problems, many service providers provide RDF Site Summary documents. In this paper, we propose a newly designed news service using RSS and VoiceXML. RSS is an XML format that supports the syndication of news stories and similar content. There are several different formats for XML syndication that are referred to as RSS. Since RSS is in XML format, turning it into VoiceXML is easy and the synergy benefits of binding RSS and VXML are great. VoiceXML is a non-proprietary, web-based markup language for creating vocal dialogues between humans and computers. In this paper, we first focus on binding RSS and VXML, and RSS feed parsing. As a result of this research, we implement enhanced real time news service. People can use our service with their wired and wireless phone at any time, at any place. Also, we validate usability by comparing a typical RSS service scenario and typical VoiceXML service scenario, and calculating user’s satisfaction.

Transport protocol in mobile devices of HCI (human-computer interaction) operates at wired and wireless heterogeneous network. it is to conquer the poor Wireless environment and a constraint mobility that have been developed from reactive congestion control schemes to proactive congestion control schemes suitable for the wired, wireless networks environment. in spite of, recently proposed that TCP’s proactive congestion control schemes could not reflect network status in detail and there are a limit to support for new mobile services of HCI users. In this paper, we outlines open issues and problems laying stress on TCP-Jersey’s available bandwidth estimators in order to reduce using rate adjustment metrics to packet loss and congestion loss. also, we presented the guideline of selecting more adequate metrics for improve wireless TCP performance of mobile HCI.

This research proposes a new task analysis methodology that combines the fuzzy Bayesian model with classic task analysis methods to develop a semi-automated task analysis tool to better help traditional task analysts identify subtasks. We hypothesize that this approach could help task analysts identify activity units performed by the call center agent. The term activity units, in our study, represent the subtasks the agents perform during a remote troubleshooting process. We also investigate whether this tool could help predict the activity units as well. An effort-intensive field-based data collection for the call center’s naturalistic decision making’s environment was accomplished. A human expert and an additional 18 Purdue students participated in the validation of the assigned subtasks. The machine learning tool’s performance was then examined. The preliminary results support our hypotheses that the fuzzy Bayesian based tool is able to learn and predict subtask categories from the agent/customer narrative telephone conversations.

This paper discusses machine learning methods and their application to Brain-Computer Interfacing. A particular focus is placed on linear classification methods which can be applied in the BCI context. Finally, we provide an overview on the Berlin-Brain Computer Interface (BBCI).

We propose a Music Recommendation System (MRS) based on GPS information and user’s mood. The system records example automatically when user selects the music, and extracts hypothesis by Inductive Logic Programming (ILP). Then the system calculates recommendation rate by the hypothesis, and recommend the music by using roulette strategy.

Improving the performance of delay-constrained video streaming over wireless networks is one of the most important issues for various multimedia-related applications, as well as for the efficient overall utilization of the wireless medium. In this paper, we propose a scheme for adaptive admission control and scheduling for frame-based multimedia traffic in the 802.11 wireless local area networks. The conventional scheduler used for IEEE 802.11e is not suitable for VBR traffic such as MPEG-4 because only a single fixed TXOP is allocated from the hybrid Coordinator. In our scheme, we divide an MPEG flow into three different classes of video-frames depending on the user-QoS, and Our scheme is distributed admission control for each divided video-frame types and poll scheduling adaptive admission control for the flow.

Our goal is to propose the optimal network model for HCI mobile terminals with limited capacity, and to provide ubiquitous services using the minimum cost and to prolong the system lifetime by reducing the energy consumption of mobile devices. In order to achieve these goals, we propose an energy-efficient routing protocol based on clustering scheme for hybrid ad hoc networks. By applying the index-based data broadcasting and selective tuning methods, the infra system performs the major operations related to clustering and routing on behalf of ad hoc nodes. The simulation result presents that our proposed scheme shows up to about 46% of performance improvements in comparison with other schemes.

This paper proposes the real-time stereoscopic conversion method with adaptable viewing distance by the viewer’s eye detection at personal stereoscopic viewing devices. The distance between a screen and viewer is often changed by moving a personal display or moving head position. It causes a visual discomfort and low quality stereoscopy. Our proposed method is divided into two parts; eye detection part to acquire viewer’s eye positions and stereoscopic generation part in real-time. In the first part, we can measure an interocular distance of a single viewer by the proposed eye detection method and also calculate the distance between a screen and a viewer by using the principle of stereoscopic generation. In the other part, we propose an efficient motion-based approach for conversion from 2D video to stereoscopic video. The disparity between left image and right image is adaptively adjusted by viewer’s interocular distance. Adaptation time for a 3D immersion during 3D stereoscopic viewing is remarkably reduced and stereoscopy is personally optimized. Therefore, our proposed system can offer optimized stereoscopic conversion and more freedom of viewer’s head considering the distance between a screen and a viewer through the proposed eye detection.

SCTP(Stream Control Transmission Protocol) is a new IETF transport protocol over the IP network. Transport protocol is responsible for reliable delivery of a message form a host to another host. We implemented an efficient file transfer system using the SCTP multiple file transfer and modified STCP congestion control mechanism for heterogeneous ubiquitous environment to improve end-to-end FTP efficiency. The result shows that proposed system reduces transmission latency for huge data and improves the performance.

In order to improve the quality of embedded software by finding obstacles unanticipated by software specifications, this paper describes a requirement analysis method based on an Information Flow Diagram and experiments using this method, with discussion. Recently, embedded software has become more large-scale and complicated while the development cycle of the software has typically been shortened. This industry trend compromises the quality of embedded software. In order to improve the quality of the software, unexpected obstacles must be carefully analyzed in the specification and design phases. We propose a diagram-based analysis method to answer this need, and have performed an experiment using the analysis method with an actual product. The results of the experiment made it clear that 1) Novice engineers can construct an Information Flow Diagram under the lead of expert engineers; and 2) The methods by which experts lead novice engineers in the analysis of unexpected obstacles require further refinement.

In this research, we have built a system for network administrators that visualize the Peer-to-Peer (P2P) file sharing activities of network users. This system monitors network traffic and discerns traffic features using traffic mining. This system visualizes the P2P file sharing traffic activities of an organization by making the processing object not an individual user but a user group. The network administrator can comprehend the P2P sharing activities of the organization by referring to the map. This system extracts a traffic feature from captured IP packets that the users communicated. Afterwards this system creates a traffic model. The features of the traffic model are emphasized by weighting. After that, the traffic model is visualized by a Self-Organizing Map. The network administrator is assisted in understanding users’ P2P file sharing communication behavior by this feature map. The administrator can then respond to the situation. As a result, we think we can assist the monitoring operation and network administration.

To deal with the problem of extracting rules from the information table, a new method was studied and proposed. First, define discernible vector and its addition rule by the indiscernible relation in rough set. Second, scan discernible vectors just only one time by the discernible vector addition rule in order to obtain the core attribute set and the important attributes. Then reduce attribute and attribute value by deleting redundant attributes and attribute values respectively. Finally, a concise rule set was obtained. The illustration and experiment results indicate that the method is effective and efficient for rule extraction.

The higher-order programming is one of the attractive and powerful ways of expressing our algorithmic/procedural methods of solving in many application fields. It represents our idea in program form naturally. The authors have studied and implemented the function of handling higher-order programming paradigm in Hyperlogo language system. The handling functions are: to make procedure closures, to handle procedure closures in the same way as numerical value and character strings, and to activate generated procedure closures. The third point of the above functions is focused mainly in this paper. Two ways of activating closures in Hyperlogo are presented: (1) add a name to generated closure, and call it up by its name as the need arises, (2) use the assistant procedure which activates a target procedure. The procedure

activate

is introduced to acquire the function.

Our aim is to implement a mind-reading capability in machines and artifacts. We give a formulation of mind-reading applicable to machines, emphasizing the importance of a human stance to a machine in mind-reading. Next we demonstrate how the stance a human takes can modify the recognition of intention in an object/agent, taking as examples experiments conducted by us, i.e., a moving box experiment, eye-contact robot, gaze-game experiment, and three-card-game experiment. To realize a human-machine communication based on mind-reading, it is essential that a human takes an intentional stance to the machine. We have to use signs of intentionality to trick a human to take an intentional stance. Shared attention is shown to work as a sign of intentionality.

Human beings are fascinating creatures. Their behavior and appearance cannot be compared with any other living organism in the world. They have two distinct features with compared to any other living being; unique physical nature and emotions / feelings. Anybody who studies on humans or trying to construct human like machines should consider these two vital facts. When robots are interacting with humans and other objects, they certainly have a safe distance between them and the object. But how can this distance be optimized when interacting with humans; will there be any advantages over achieving this; will it help to improve the condition of robots; can it be a mere constant distance; how will the humans react, are some questions arosed. In order to “humanize” robots, they (robots) should also have certain understating of such emotions that we, humans have. In this research project, authors are trying to “teach” one such human understanding, commonly known as “personal space” to autonomous mobile robots.

This study explored the feasibility of displaying dynamic Chinese text on the screen of a wrist watch. Three design factors (i.e., dynamic display, presentation method, and speed) were examined to investigate their effects on users’ reading comprehension under two different conditions of task types. The results of this study indicated the following: (1) There was no significant difference between leading display and RSVP in both single- and dual-task conditions; (2) Presentation method was a significant factor. The participant’s reading comprehension was significantly better with the word-by-word format than with the character-by-character one in both single- and dual-task conditions; (3) The participant’s reading comprehension was not significantly different among the three different speed settings in the single-task condition. However, participants had significantly higher reading comprehension scores under the slower speed settings of 150 and 250 cpm than under the faster speed setting of 350 cpm in the dual-task condition.

Walking with someone or guiding them to a destination is a simple task that humans do everyday. However, while almost any mobile robot can navigate to a given point while a human walks behind it, these robots do not take into consideration whether the human is following along properly. Our research involves the use of a wireless sensor network for navigation of a guidance robot and a vision-based tracking system for human awareness. The network creates a virtual directional field that provides a directional imperative to the robot, and the vision system allows it to detect if a following human has strayed. We also present a demonstration of our system in a real world application.

This paper studies an elemental human action, hand up-down, and selective button operations, by means of NIRS. In the analysis, we pay attention to NIRS trajectory and the total Hb changes. Our results provided empirical support for the hypothesis that the trajectory of positive inclination indicates brain location actively contributing the task, and negative inclination is the brain locations out of contribution. The locations positively contributing to the task are found most intensively in right and left forebrain area but in middle part of left brain looks less contributing to the task.

This paper presents an interactive design project of a remote clinical robot for ophthalmology, for use in support of telemedicine. The project is a case study of finding the best organizational structure of a Graphical User Interface (GUI) and Solid User Interface (SUI) in the age of ubiquitous computing. At the early stage of design, we carried out usability tests with paper-based prototypes and obtained requirements from ophthalmologists. Results showed that paper prototyping is applicable for GUI- and SUI-combined products, even they have manual control input and continuously changing visual feedback. We examine lessons learned from the design process and designed product.

The human’s facial expression plays an important role as media that visually transmit feelings and the intention. To express the facial expression freely, the action of muscles of facial expression forms the expression is very important. “Facial expression training” is recently paid to attention, and is a training of the face to make a good facial expression. There is a training method by moving the expression muscle of the face. One of the purposes to train the facial expression is to achieve your own ideal facial expression. In this study, it aims at a support of the effective process for facial expression training to achieve the target expression using computer, as a self training system. And, an interface for users to select a target facial expression and a whole development of an effective expression training system is proposed, as a first step toward an effective facial expression training system.

As robots become more intelligent and their application fields continue to grow, the decisions and interaction of robots that share work domains with humans become increasingly important. Traditional robots have received only simple commands, and humans’ roles have been limited to supervisor. However, for successive task performance, robots’ decision making should be approached via collaboration between the human and robot. Interaction also should be regarded as an issue closely associated with joint work plans rather than a simple function. Interaction between the human and robot, moreover, should be systemized in order to decrease the workload of the human and maximize user satisfaction. Accordingly, we developed several cognitive models: a task model, truth-maintenance model, interaction model, and intention rule-base. These models can manage and initiate the interactions based on their tasks and modify robot’s activities by using the result of the interaction. We demonstrated the adaptability and usability of the developed models by applying them to the home-service robot, T-Rot.

In this paper, A stereoscopic display system using a rotary disk type beam shutter and two beam projectors is proposed. It is a kind of active stereoscopic display and can be easily converted from a passive stereoscopic system using polarizing filters. If it is possible to synchronize the revolution speed of the beam shutter with the integer multiple of the scanning frequency of the beam projectors, we can obtain a comparable performance with our system to an active stereoscopic system using one expensive high performance beam projector. Further, if we rotate the beam shutter at sufficiently high revolution speed, our active stereoscopic system works regardless of synchronization and thus the system is much easier to implement.

We clarify internal timing control mechanism in a synchronization tapping task. Synchronization error (SE) was directly controlled within 450–1800 ms of the inter stimulus-onset interval (ISI) with our experimental method. Two types of internal phase correction mechanism became evident. The first showed a strong negative correlation between SE change and change in inter tap-onset interval (ITI), suggesting a simple negative feedback mechanism. This correlation was observed under all ISI conditions. The second type showed a larger response with a low negative correlation and was observed only with long (1200–1800 ms) ISIs. Only the second mechanism was inhibited in a dual task condition. These results suggest that there are two types of internal timing mechanism.

We examined brain activation in the frontal cortex during the contents understanding of comic stories with near-infrared spectroscopy (NIRS). The present results showed that the left prefrontal lobe region is activated when people actively try to understand the comic stories and to memorize their contents for reporting in the future. The activated area could overlap with or be around Broca’s area. We discussed how to apply this brain activation related to human understanding to evaluation of contents presented in the human-machine interfaces.

Using NIRS, we investigated brain activity in subjects searching for target characters in given text. Visual stimuli consisted of two different contrasts: positive and negative. Both stimuli were displayed for 30 sec on a cathode-ray tube (CRT: 21 inches). The subjects were instructed to watch a visual display on the CRT. This display included text without meaning. When target characters were detected, subjects were instructed to perform three tasks: (1) count the number of target characters by visual observation, (2) tap a button with the thumb and fingers of their left hand carefully and as fast as possible and (3) operate a mobile phone.

Intelligent robots can obtain environmental information by processing sensor signals, understand contextual information, and perform necessary tasks autonomously. They can also perform tasks directed by commands from a user. They can be used for housekeeping, entertaining, and educational purposes. In order to use intelligent robots efficiently, a convenient user interface is needed to control them. In this study, we analyze how to organize such a user interface according to the function and level of autonomy of intelligent robots and develop a user interface for controlling them remotely via a PDA (Personal Digital Assistant) or mobile phone.

This paper discusses development of a control system for home robots. When a user requests a user interface for home robot, the home robot sends this robot’s user interface to the user’s PDA. Then, the robot can be efficiently controlled from this user interface on the PDA. The home server distributes the service of robots, and manages the transmission of service interface modules. When this procedure is used, home network traffic is reduced, and the user can simply access a service of a robot.

The increasing capabilities of experimental household robot platforms require more and more sophisticated methods of interaction. While there are many developments in all directions of Human-Machine-Interaction, the integration and combination of several modalities into one robot system require some effort. To ease the development of applications supporting several types of interaction, Fraunhofer IPA has developed a framework named “Go”. Within this framework we have integrated different kinds of interaction methods into one robot platform “Care-O-bot 3”, a mobile service robot for accomplishing daily tasks. This framework and its interaction methods are presented here.

To quantitatively clarify the cognitive mechanism of KANSEI information conveyed by body movement, we conducted evaluation experiments and movement analysis by optical motion capture. We attempted to develop multiple regression models to investigate the effect of physical movement on subjective evaluation. The results show the relation between subjective feature of movements and the physical measurement. Thus, we expect that our research is effective to the guidance of dancing and acting, because it is possible to paraphrase sensitive image (KANSEI information) in another word (movement feature) which has strong relation with physical movement. Furthermore, this relational model which can replace human sensibility by digital data is useful for human-computer interaction.

People often forget how to write Kanji characters even though they are able to read those characters. If one cannot write a Kanji character, one cannot input it with current handwriting recognition engines. Therefore, we propose to combine Kana (phonetic characters) to Kanji conversion with a handwriting recognition engine. Moreover, we remove writing boxes to write characters one by one, since they are unnatural for many applications. This is similar to writing on a plain paper with a pen. When, handwritten characters are misrecognized, the user corrects them using correction gestures. When the user cannot write Kanji characters, the user writes in Kana and invokes the Kana-to-Kanji conversion. In reality, people often write Kana characters instead of Kanji characters when they cannot write those characters or it takes more time than writing Kana characters. Kana-to-Kanji conversation is common in keyboard interface. This paper describes the design of a pen interface by which the user can write text without writing boxes, correct misrecognitions and specify Kana-to Kanji conversion. When a user writes Kana characters instead of Kanji characters, it would be desirable to translate them to Kanji characters automatically.

This paper describes the tasting robot with an artificial sense of taste. Its optical tongue examines the taste of food and drink within several seconds and gives the name as well as its ingredients without crashing them and without opening paper or plastic packages. It is the integration of infrared (IR) spectroscopic technologies with pattern recognition technologies. The robot is the world’s first tasting robot which gives advices on the food and on health issues based on the information gathered. It sees if foods are best to eat or if they are good for health from the point of the content of sugar and fat. The robot was first exhibited in EXPO 2005 AICHI, Japan.

This paper describes a concept-based approach for question answering system in which concept rather than keyword makes an important role on question analysis, document retrieval, and answer extraction. Our idea is that we can extract correct answer from various paragraphs with different structures when we use well-defined concepts because concepts occurred in questions of same answer type are similar. We defined a concept rule for each answer type. The concept rule contains core concepts occurred in questions of that answer type. Question analysis module extracts concepts from user’s question and determines the answer type. Document retrieval module retrieves more relevant documents using extracted concepts. Answer extraction module extracts a probable answer from retrieved documents using concepts. Empirical results show that our concept-based approach can retrieve more relevant documents and extract more accurate answer than any other conventional approach. Also, our approach has additional merits that it is language universal model, and can be combined with arbitrary conventional approaches.

There are needs recently to push information technology beyond the desktop environment. It is because even decades after the proliferation of information technology, it still, by large, confine the users in its own domain. Conventional IT confined users to the desktop and virtual reality technology confined the users onto specialiced devices such as the HMD, large screen display room, or the CAVE. Our vision is to push for the ubiquity of IT beyond the mobile cell phone media. This paper discusses several requirements for the real IT vision. Spefically we are interested in the characteristics of display technologies that suit our vision. We will describe several key technologies that have been developed within this requirements, and our most recent work in this field.

This paper gives an overview of recent projects done at Fraunhofer IAO which take innovative approaches to the use of non-speech audio in human-computer interfaces. The examples illustrate the benefits that non-speech sound - alone and in combination with other modalities - can provide for supporting an effective interaction.

An approach to haptic interaction and simulation of motion and deformation of elastic object is proposed. In our approach, the motion of an object is defined by changes of kinetic momentum and angular momentum after impulse force is applied to each degree of freedom, and the deformation of the object is defined by a set of temporal deformation patterns after impulse force is applied to each degree of freedom, which we call extended impulse response deformation model. The time complexity of computing interaction force is independent of the complexity of the model. This feature is advantageous for time-critical applications. Also, the time complexity of computing object deformation is linearly proportional to the complexity of the model. Through implementation of a prototype environment and evaluation of its performance, the feasibility of the proposed approach is demonstrated.

NIRS trajectory method is proposed to understand brain activities engaged in mobile device operation. The trajectory is the plot of oxy- and deoxy- Hb (hemoglobin concentration) data to the Oxy-Deoxy Hb plane. Three basic trajectories and some complex trajectories are explained. Total brain activities are represented by a set of NIRS trajectories derived from NIRS channels located on the head surface. Trajectory maps of text entry to mobile phone are shown. The trajectory map increases its complexity with the increase of work task complexity, which we defined by normal and reverse order hiragana and kanji conversion entry of Japanese text. The results suggest the right brain is taking dominant roles in text and other finger activities. NIRS trajectory method can be applied further to find centers of brain activities located on the surface area of the brain. Kindly use color version in CD-ROM for better understanding of this paper.

The electroencephalogram signals of steady-state visual evoked potentials were recorded for three subjects in immersive virtual environment. A machine learning technique of support vector machine with single trial EEG data for 1.0 seconds resulted in 92.1 % of averaged recognition rate in selecting a virtual button among two. The online demonstrations in CAVE showed good performance in position control of a simple 3D object.

The embodied communication system for mind connection (E-COSMIC) has been developed by applying the entrainment mechanism of the embodied rhythms of nodding and body movements to physical robots and CG characters. E-COSMIC comprises an embodied virtual communication system for human interaction analysis by synthesis and a speech-driven embodied interaction system for supporting essential human interaction and communication based on the analysis that uses the embodied virtual communication system. Human-entrained embodied interaction and communication technology for an advanced media society is introduced through some applications of E-COSMIC.

InterRobot is an embodied interaction robot driven by InterRobot technology (iRT) in order to make communicative motions and actions based only on speech input. In this paper, the concept of an embodied image telecasting method is proposed for activating embodied interaction and communication via such a robot. The prototype systems are developed: a new InterRobot equipped with a camera and a projector, and an embodied visual effect on telecasted images. The sensory evaluation demonstrates that the visual effect is effective for realizing presence and a sense of unity.