Skip to main content

Über dieses Buch

This book constitutes the refereed proceedings of the 7th International Conference on Active Media Technology, AMT 2011, held in Lanzhou, China, in September 2011. The 30 revised full papers and 6 keynote talks were carefully reviewed and selected for inclusion in the book. They are grouped in topcial sections on data mining and pattern analysis in active media; active human-Web interaction and social media; active Web intelligence applications; active multi-agent and network systems; as well as technology intelligence.



Keynote Talks

People’s Opinion, People’s Nexus, People’s Security and Computational Intelligence: The Evolution Continues

The talk begins with a brief introduction to some of our research work in the past few years as well as the ongoing research. A new model on extending the flexibility and responsiveness of websites through automated learning for customtailoring and adaptive web to user usage patterns, interests, goals, knowledge and preferences will be presented. The second part of the talk will be devoted to the challenges that the Computational Intelligence communities are faced with in order to address issues related to people’s nexus, opinion, and security on the Web, and our contributions to these topics. At the end, I will provide an overview of our current research focus on network security and intelligence information handling and disimination.

Ali Ghorbani

Towards Conversational Artifacts

Conversation is a natural and powerful means of communication for people to collaboratively create and share information. People are skillful in expressing meaning by coordinating multiple modalities, interpreting utterances by integrating partial cues, and aligning their behavior to pursuing joint projects in conversation. A big challenge is to build conversational artifacts – such as intelligent virtual agents or conversational robots – that can participate in conversation so as to mediate the knowledge process in a community. In this article, I present an approach to building conversational artifacts. Firstly, I will highlight an immersive WOZ environment called ICIE (Immersive Collaborative Interaction Environment) that is designed to obtain detailed quantitative data about human-artifact interaction. Secondly, I will overview a suite of learning algorithms for enabling our robot to build and revise a competence of communication as a result of observation and experience. Thirdly, I will argue how conversational artifacts might be used to help people work together in multi-cultural knowledge creation environments.

Toyoaki Nishida

The Global-First Topological Definition of Perceptual Objects, and Its Neural Correlation in Anterior Temporal Lobe

What is a perceptual object? This question seems to be straightforward yet its answer has become one of the most central and also controversial issues in many areas of cognitive sciences.

The “global-first” topological approach ties a formal definition of perceptual objects to invariance over topological transformation, and the core intuitive notion of a perceptual object - the holistic identity preserved over shape-changing transformations - may be precisely characterized as topological invariants, such as connectivity and holes.

Lin Chen, Ke Zhou, Wenli Qian, Qianli Meng

Combinatorial Fusion Analysis in Brain Informatics: Gender Variation in Facial Attractiveness Judgment

Information processing in the brain or other decision making systems, such as in multimedia, involves fusion of information from multiple sensors, sources, and systems at the data, feature or decision level. Combinatorial Fusion Analysis (CFA), a recently developed information fusion paradigm, uses a combinatorial method to model the decision space and the Rank-Score Characteristic (RSC) function to measure cognitive diversity. In this paper, we first introduce CFA and its practice in a variety of application domains such as computer vision and target tracking, information retrieval and Internet search, and virtual screening and drug discovery. We then apply CFA to investigate gender variation in facial attractiveness judgment on three tasks: liking, beauty and mentalization using RSC function. It is demonstrated that the RSC function is useful in the differentiation of gender variation and task judgment, and hence can be used to complement the notion of correlation which is widely used in statistical decision making. In addition, it is shown that CFA is a viable approach to deal with various issues and problems in brain informatics.

D. Frank Hsu, Takehito Ito, Christina Schweikert, Tetsuya Matsuda, Shinsuke Shimojo

Study of System Intuition by Noetic Science Founded by QIAN Xuesen

This talk investigates the meaning, contents and characteristics of systems institution on the basis of Noetic Science, which was founded by Qian Xuesen. The systems intuition is the human capability to find the hidden system imagery of the object or to create an imagery of new system. The basic noetic foundation of system intuition and cultural influence to it are studied. The open problems are also listed.

Zhongtuo Wang

Study of Problem Solving Following Herbert Simon

Herbert Simon (1916.6.15 - 2001.2.9) was one of the greatest pioneers in cognitive science and artificial intelligence, as well as in behavior economics and many other fields. Problem solving was his core work in artificial intelligence and cognitive psychology. He and Newell first postulated a general and systematic framework of human (and machine) problem solving as iteratively applying operators to transform the state of the problem from the starting state in problem state space to eventually achieve the goal state. Heuristic problem solving includes two basic components: heuristic searching (such as means-ends analysis) and heuristic rules (used to change the problem states). And then, he extended this framework in two dimensions. One is applying this framework to creative learning and scientific discovery (both were thought as specific ill-structured problem solving tasks); the other is to elaborate this general framework with more detailed models in memory (such as chunk structure in short term memory) and the knowledge (and problem) representations, including the knowledge structure difference between experts and naives, diagrammatic representation and mental imagery. To meet the challenge of Web intelligence and to pioneer the effective and efficient ways of information processing at Web scale, as the first step, we would learn this process from human brain, one of the greatest webs, based on Simon and Newell’s framework in problem solving. We have found that, even in the basic application of heuristic rules, the processes are distributed in several major parts of brain and with certain areas for the communications across these networks. We have checked the brain activations in regard to working memory and mental imagery in problem solving. We have also found the evidences supporting the hypothesis that the scientific discovery is a specific problem solving from neural activations that central brain areas activated in scientific discovery overlapping with the areas in general problem solving tasks. These findings offer strong clues for how to solve problems at Web scale.

Yulin Qin, Ning Zhong

Data Mining and Pattern Analysis in Active Media

A Heuristic Classifier Ensemble for Huge Datasets

This paper proposes a heuristic classifier ensemble to improve the performance of learning in multiclass problems. Although the more accurate classifier leads to a better performance, there is another approach to use many inaccurate classifiers while each one is specialized for a few data in the problem space and using their consensus vote as the classifier. In this paper, some ensembles of classifiers are first created. The classifiers of each of these ensembles jointly work using majority weighting votes. The results of these ensembles are combined to decide the final vote in a weighted manner. Finally the outputs of these ensembles are heuristically aggregated. The proposed framework is evaluated on a very large scale Persian digit handwritten dataset and the experimental results show the effectiveness of the algorithm.

Hamid Parvin, Behrouz Minaei, Hosein Alizadeh

Ontology Extraction and Integration from Semi-structured Data

Domain ontologies are usually built by domain expert manually. They are accurate and professional from the perspective of domain dependent concepts, instances and relations among them, nevertheless, maintaining and creating new ontologies need too much manual work, especially when the ontology goes to large scale. Semi-structured data usually contain some semantic relations for concepts and instances, and there are many domain ontologies implicitly exist in these types of data sources. In this paper, we investigate automatic hierarchical domain ontology generation from semi-structured data, more specifically, from HTML and XML documents. The main process of our work includes domain terms extraction, pruning, union and hierarchical structure representation. We illustrate our study based on Artificial Intelligence related conference data represented in HTML and XML documents.

Shaobo Wang, Yi Zeng, Ning Zhong

Effectiveness of Video Ontology in Query by Example Approach

In this paper, we develop a video retrieval method based on Query-By-Example (QBE) approach where a query is represented by providing example shots. Relevant shots to the query are then retrieved by constructing a retrieval model from example shots. However, one drawback of QBE is that a user can only provide a small number of example shots, while each shot is generally represented by a high-dimensional feature. In such a case, a retrieval model tends to be overfit to feature dimensions which are specific to example shots, but are ineffective for retrieving relevant shots. As a result, many clearly irrelevant shots are retrieved. To overcome this, we construct a

video ontology

as a knowledge base for QBE-based video retrieval. Specifically, our video ontology is used to select concepts related to a query. Then, irrelevant shots are filtered by referring to recognition results of objects corresponding to selected concepts. Lastly, QBE-based video retrieval is performed on the remaining shots to obtain a final retrieval result. The effectiveness of our video ontology is tested on TRECVID 2009 video data.

Kimiaki Shirahama, Kuniaki Uehara

A Survey of Energy Conservation, Routing and Coverage in Wireless Sensor Networks

The emergence of multimedia wireless sensor networks and its characteristics bring in new problems of wireless sensor networks compared with the traditional networks and wireless networks. We mainly consider three issues in this paper, energy conservation, coverage and efficient routing, all of which are fundamental in designing and implementation of wireless sensor networks. Energy conservation is the most significant problem in wireless sensor networks due to its limited energy source intuitively. The other two issues are usually combined with the energy conservation problem. The goal of coverage requirement is to have each location in the targeted physical space within sensing range of at least one sensor node. Besides, efficient routing aims to solve the problem that the collected data are efficiently reported to end-users. We introduce several state-of-the-art works and conclude these researches concerning their various metrics. Finally, we propose some rules while designing the wireless sensor networks according to the previous works, concerning these metrics.

Wang Bin, Li Wenxin, Li Liu

A Multi-type Indexing CBVR System Constructed with MPEG-7 Visual Features

Since multimedia has played an important role in our daily life, multimedia mining becomes a popular research area. Among the emerging research topics of multimedia mining, content-based video retrieval is a challenging one which attracts researchers’ attention. Here, we make an attempt to retrieve the requested video contents from a long-length video. We issue one

semantic query clip

extracted from a source video, and try to retrieve the requested video contents of this source video. In this paper, we proposed a novel scheme to facilitate indexing and/or querying long-length videos. For indexing source videos, we construct and integrate two index structures:






. Thus, we can easily locate candidates from the index structures and perform sequence matching to retrieve the requested video contents. They could be the video contents exactly matched with or similar to the query clip. Finally, the experimental results show that the proposed scheme can retrieve requested video contents in an effective way.

Yin-Fu Huang, He-Wen Chen

A Novel Data Collection Scheme Based on Active Degree for OMSN

Opportunistic mobile sensor networks manifest profoundly practical values. Its sink nodes have no fixed connection with the source node, which adopt “store-carry-forward” paradigm to transmit data during random meetings. On the circumstance, we propose a novel data collection scheme and a forwarding strategy based on Active Degree (AD). This scheme calculates the AD from each node to all sink nodes based on a definition of expected latency and the concept of entropy. Messages are forwarded according to the AD values in the degrading order during the occasion, until the message reaches the sink node. Simulation experiments reveal that the scheme is able to adapt to networks’ dynamic evolution, which is robust, and has low cost of calculation and storage, which is adaptable to sensor nodes with limited resources.

Jianwei Niu, Bin Dai, Jinkai Guo

Research of Robust Facial Expression Recognition under Facial Occlusion Condition

Robust facial expression recognition under facial occlusion condition is the main research orientation, which has important research significance. Many problems are caused by facial occlusion, not only missing facial expression information, but also bringing outliers or lots of noise. Aiming at the point, firstly, the face to be recognized is reconstructed using robust principal component analysis (RPCA); secondly, Eigenfaces and Fisherfaces are used to extract facial expression features respectively; finally, nearest neighbor method and support vector machine are used as classifiers. Facial expression recognition experiments are implemented in different occlusion conditions on Japanese female facial expression database (JAFFE). On the condition of big occlusion and small sample, RPCA algorithms gained better recognition results than many other methods, showing that this method based on RPCA is robust to kinds of facial occlusions.

Bin Jiang, Ke-bin Jia

Active Human-Web Interaction and Social Media

Visualizing Secure Hash Algorithm (SHA-1) on the Web

Cryptographic hash functions play an important role in modern cryptography. In this paper, we present web-based interactive visualization applets for teaching secure hash algorithm (SHA-1). The visualization applets lead to more efficient learning outcomes. The visualization applets of SHA-1 contain description of SHA-1, animation, and controls for activation with SHA-1. The primary goal of this visualization is to understand the essence of SHA-1 and to be able to find out the strength and weakness of SHA-1. The applets have been used in our department for the course “cryptography” for master students. Most of the students agreed that the applets are helpful to understand SHA-1 and some of its vulnerabilities. The applets can be generalized to SHA-2 and other hash functions.

Dalia B. Nasr, Hatem M. Bahig, Sameh S. Daoud

Emotion and Rationality in Web Information: An Eye-Tracking Study

Web pages are the main interface of human-computer interaction on the Internet. Although several factors of successful Web page design have been studied, few researches considered the utility and effectiveness of emotion and rationality in Web information for e-Commerce. This study conducted an eye-tracking experiment to address how emotional and rational appeals mediate the users’ Web commercial information perceiving, as reflected by their own eye movements. The major finding is that the mean fixation duration for participants to obtain information appealed in emotional strategy is significant shorter than those appealed in rational strategy. The result may indicate that Web commercial information in rational strategy is more difficult to be extracted, and therefore suffers from more cognitive processes.

Linchan Qin, Ning Zhong, Shengfu Lu, Mi Li, Yangyang Song

Constructing the Internet Behavior Ontology: Projection from Psychological Phenomena with Qualitative and Quantitative Methods

The Internet has intrigued new research paradigm for traditional psychology. Over the past decades, researchers have found that there exists strong correlation between Internet behavior and individual psychological characteristics. In this paper, we propose to build the ontology of Internet behavior by projecting from psychological phenomena. The psychological phenomenon index was developed with top-down literature study. We conduct focus group study and build an online open-ended questionnaire to collect the comprehensive Internet behaviors as much as possible according to the psychological index. This study combines the qualitative and quantitative methods and sets the foundation for further study via the Web.

Qi Zhang, Zhuo-Hong Zhu, Ting-Shao Zhu, Jiu-Ling Xin, Shu-Juan Wang, Wei-Chen Zhang, Ang Li, Yi-Lin Li, Shan Tang, Yu-Xi Pei

Why Do People Share News in Social Media?

Sharing news in social media has influence on individuals as well as society and has become a global phenomenon. However, little empirical research has been conducted to understand why people share news in social media. Adopting the uses and gratifications theory, we investigate the gratification factors influencing news sharing intention on social media. A regression analysis was employed to analyze the data collected from 203 undergraduate and graduate students. The results show that informativeness was the strongest motivation in predicting news sharing intention, followed by socializing and status seeking. However, entertainment/escapism was not a significant predictor in contrast to prior work. Implications and opportunities for future work are also discussed.

Chei Sian Lee, Long Ma, Dion Hoe-Lian Goh

Active Web Intelligence Applications

Hot Topic Detection in Professional Blogs

Topics in professional blogs mainly refer to specific techniques. Today, professional blog websites have been important information sources. However, information overload and the uncertainty of topic hotness evaluation have been obstacles for hot topic detection. The paper proposes a method of detecting hot topics in professional blogs. The proposed method is based on the characteristics of the professional blogs and mainly analyzes candidate topics that are likely to be hot. First, a word network based on high frequency keywords and co-occurrences of the keywords is constructed, and then the candidate topics are extracted by analyzing the structure of the word network. The opinion networks with respect to the topics in different time intervals are subsequently constructed for opinion analysis. Finally, hot topics are identified by computing the user participation degree, opinion communication degree, and timeliness of the candidate topics. Experimental results show the proposed method is feasible and reasonable.

Erzhong Zhou, Ning Zhong, Yuefeng Li

A Weighted Multi-factor Algorithm for Microblog Search

As a fast and social information communication media, microblog, especially Twitter, has gained increasing popularity in recent years. Given the fact that a great volume of new tweets are being generated every second, ranking them to find the most relevant information is a challenging matter. The short length of tweets makes direct adoptions of traditional information retrieval algorithms to microblog search very hard. In this paper, we focus on the ranking strategies of microblogs, six factors are summarized to measure a user’s social influence, and each of them are highly relevant to the social network properties of the microblog authors and the properties of the microblog itself. Based on these factors, several ranking measures for Twitter search are examined. As a step forward, we propose a weighted multi-factor ranking algorithm (WMFR). By using a public Twitter search dataset, through Kendall’s


correlation analysis on user selection and algorithm selection of tweets, we conclude that the proposed WMFR algorithm is more effective compared to several existing algorithms.

Lulin Zhao, Yi Zeng, Ning Zhong

A Combination Ranking Model for Research Paper Social Bookmarking Systems

Social bookmarking systems are essential tools for web resource discovery. The performance and capabilities of search results from research paper bookmarking system are vital. This paper proposes a combination of similarity based indexing “tag title and abstract” and static ranking to improve search results. In this particular study, the year of the published paper is combined with similarity ranking called (


). Different weighting scores are employed. The retrieval performance of these weighted combination rankings are evaluated using mean values of NDCG. The results indicate that


and similarity rank with weight 90:10 has the highest NDCG scores. The result from the experiments implies that the chosen heuristic ranking may improve the efficiency of research paper searching on social bookmarking websites.

Pijitra Jomsri, Siripun Sanguansintukul, Worasit Choochaiwattana

An Upgrading Feature-Based Opinion Mining Model on Vietnamese Product Reviews

Feature-based opinion mining and summarizing (FOMS) of reviews is an interesting issue in the opinion mining field. SentiWordNet is an useful lexical resource for opinion mining, especially for FOMS. In this paper, an upgrading FOMS model on Vietnamese reviews on mobile phone products is described. Feature words and opinion words were extracted based on some Vietnamese syntactic rules. Extracted feature words were grouped by using HAC clustering and semi-supervised SVM-kNN classification. Customers’ opinion orientation and summarization on features was determined by using a VietSentiWordNet, which had been extended from an initial VietSentiWordNet. Experiments on feature extraction and opinion summarization on features are showed.

Quang-Thuy Ha, Tien-Thanh Vu, Huyen-Trang Pham, Cong-To Luu

Predicting Mental Health Status Based on Web Usage Behavior

To build a predicting model for mental health status based on Web Usage Behavior, we collect data from 571 first-year graduate students using our own Internet Usage Behavior Check-List (IUBCL) and Psychological Health Inventory (PHI). We build six logistic regression models, in which Web usage behavior features are as independent variables while mental health status as dependent ones. We find that the accuracy is about 72.9% − 83.1%, which demonstrates it is applicable and feasible to identify each individual’s mental health status by analyzing his/her Web usage behaviors.

Tingshao Zhu, Ang Li, Yue Ning, Zengda Guan

User Interests Modeling Based on Multi-source Personal Information Fusion and Semantic Reasoning

User interests are usually distributed in different systems on the Web. Traditional user interest modeling methods are not designed for integrating and analyzing interests from multiple sources, hence, they are not very effective for obtaining comparatively complete description of user interests in the distributed environment. In addition, previous studies concentrate on the text level analysis of user interests, while semantic relationships among interests are not fully investigated. This might cause incomplete and incorrect understanding of the discovered interests, especially when interests are from multiple sources. In this paper, we propose an approach of user interest modeling based on multi-source personal information fusion and semantic reasoning. We give different fusion strategies for interest data from multiple sources. Further more, we investigate the semantic relationship between users’ explicit interests and implicit interests by reasoning through concept granularity. Semantic relatedness among interests are also briefly illustrated for information fusion. Illustrative examples based on multiple sources on the Web (e.g. microblog system Twitter, social network sites Facebook and LinkedIn, personal homepage, etc.) show that proposed approach is potentially effective.

Yunfei Ma, Yi Zeng, Xu Ren, Ning Zhong

Tags Weighting Based on User Profile

The ’Collaborative Tagging’ is gaining popularity on Web 2.0, this new generation of Web which makes user reader/writer. The ’Tagging’ is a mean for users to express themselves freely through additions of label called ’Tags’ to shared resources. One of the problems encountered in current tagging systems is to define the most appropriate tag for a resource. Tags are typically listed in order of popularity, as del-icio-us. But the popularity of the tag does not always reflect its importance and representativeness for the resource to which it is associated. Starting from the assumptions that the same tag for a resource can take different meanings for different users, and a tag from a knowledgeable user would be more important than a tag from a novice user, we propose an approach for weighting resource’s tags based on user profile. For this we define a user model for his integration in the tag weight calculation and a formula for this calculation, based on three factors namely the user, the degree of approximation between his interest centers and the resource field, expertise and personal assessment for tags associated to the resource. A resource descriptor containing the best tags is created.

Saida Kichou, Hakima Mellah, Youssef Amghar, Fouad Dahak

A Context-Aware Recommender System for M-Commerce Applications

M-commerce is an attractive research area due to its relative novelty, rapid growth, and great potential in business applications. However, the development of M-commerce applications is facing with some physical constraints of mobile devices and barriers of existing execution models. Moreover, the nomadic users might consume enormous time to search for satisfactory products or services from abundant options with the limited capability of physical devices. Therefore, a sophisticated recommendation algorithm which attempts to recommend a list of user-preferred products or services should be incorporated in M-commerce applications. In this paper, we propose a personalized Context-aware M-commerce Recommender System which exploits the advantages of collaborative filtering and common understanding of contextual information. Since the recommendation algorithm is embedded in a layered system and closed related with other system components, we will present a comprehensive framework to integrate the concepts of mobile agent, ontology-based context model as well as service discovery and selection mechanism. We have developed a prototype to evaluate the feasibility and effectiveness of our proposal.

Jiazao Lin, Xining Li, Yi Yang, Li Liu, Wenqiang Guo, Xin Li, Lian Li

Towards Coequal Authorization for Dynamic Collaboration

In dynamic collaboration, participants oftentimes need to share resources with each other under the same criteria. However, since each participant has its own authorization policies as a way of controlling resource access, their discrepancies make such collaboration difficult. It is desired to develop a practical and automatic way to generate the collaborative policies for

coequal authorizations

. In this paper, we investigate this problem by proposing an authorization framework based on the widely adopted XACML policy. Each practical XACML policy is converted into Boolean expressions and further refined as a set of atomic rules against the policy structure. With the rule set, the combination algorithms in policies and the collaboration preference of participants, the collaborative authorization policy is automatically generated. We analyze the consistency of the collaborative policies with previous authorization policies. Some experiments are performed to exam our approach and show that it can efficiently solve the problem of coequal authorizations.

Yuqing Sun, Chen Chen

Active Multi-Agent and Network Systems

Programming Large-Scale Multi-Agent Systems Based on Organization Metaphor

Modern software systems show some characteristics (e.g., adaptation, self-organization, etc.) as the human organizations and society. In the literature of agent-oriented software engineering, organization metaphor is adopted to manage the complexity of large-scale multi-agent systems (MAS), but the potential is not entirely exploited due to a lack of explicit organizational concepts in programming languages and execution infrastructure. This paper investigates the properties and requirements to develop large-scale MAS, and proposes a new programming model by integrating organization theory into agent technology. The approach takes both organizations and roles as first-class programming entities. An enactment mechanism based on roles is proposed to compose the system, which postpones the software composition from design time to runtime to provide flexibility and dynamic. The implementation issues are discussed and a case is studied lastly.

Cuiyun Hu, Xinjun Mao, Yuekun Sun, Huiping Zhou

A Framework for Context-Aware Digital Signage

A framework for building and operating context-aware multimedia content on digital signage in public/private spaces is presented. It enables active and multimedia content to be composed of mobile agents, which are self-contained programmable entities that can travel from computer to computer and provide multimedia content for advertising or user-assistant services to users by using their own programs. The framework automatically deploys their agents at computers close to their current positions to provide advertising or annotations on objects or users. To demonstrate the utility of the framework, we present a user-assistant that enables shopping with digital signage.

Ichiro Satoh

EMTAN: A Web-Based Multi-Agent System Architecture for Input Automation

Many enterprises and factories have been using standard Web-based procedures as the centralized interface for remote control and manufacturing process data input, which is called “Web-based digital dashboard”. With the Web interface, people need to fill in data manually and then submit entire data to a remote system to process control procedure. In this paper, we have used several creativity methods to develop a multi-agent system, namely EMTAN, which can collect all kinds of data from the on-line database, electronic files, and controller’s interface. This system can also automatically retrieve all HTML fields and transfer data into the Web-based digital dashboard, and generate the results via using email or mobile instant messages. Our multi-agent system, in fact, is an integrated model to help process Web-based data precisely and can save lots of time on data key in.

Ming-Jui Huang, Cheng-Tao Chuang, Kai-Hsiang Yang, Cheng-Yuan Liou

Estimating the Density of Brown Plant Hoppers from a Light-Traps Network Based on Unit Disk Graph

This paper is aimed at introducing a new approach to estimate the density of Brown Plant Hoppers (BPHs) at provincial scale. The model is based on the topology of a light-traps network (to gather the information about the BPHs) of a province. The BPHs density is determined based on Unit Disk Graph technique where each light-trap becomes a vertex and the edges reflect the relations on the mutual transfer of BPHs between light-traps. The model uses the historical light-traps data as the input to estimate the density of unknown location via an influence function. The experimental results of the model are performed in a typical province of the Mekong Delta region, namely Dong Thap province of Vietnam.

Viet Xuan Truong, Hiep Xuan Huynh, Minh Ngoc Le, Alexis Drogoul

Modelling the Behaviour of Crowds in Panicked Conditions

Recently there has been an increasing amount of research being done on the topic of swarm intelligence and modelling crowd behaviour. The reasons for this being that a variety of disciplines find simulations of swarm intelligence to be very useful. There is however one area in particular in which more research could be done, the modelling of crowds as they disperse. The dispersal of crowds can occur for several reasons, such as naturally dispersing as a crowd leaves a confined area, or because of perceived dangers suddenly appearing within the crowd which can cause panic and confusion. This work is focused on the later of these causes. Specifically, this paper will examine how modelling groups within a crowd can effect the simulation of crowds dispersing. We were able to determine that by incorporating group behaviour into our model it did have an impact on many factors of the simulation, including the time it took for agents to escape. By taking into account such group structures within crowds we believe future simulations can be made much more accurate.

Jake Wendt, Guangzhi Qu, Jianwei Niu

How to Play Well in Non-zero Sum Games: Some Lessons from Generalized Traveler’s Dilemma

We are interested in two-person games whose structure is far from zero-sum. We study the iterated

Traveler’s Dilemma

(TD) which is a two-player, non-zero sum game that, depending on the exact values of its critical parameters, may offer plenty of incentives for cooperation. We first briefly summarize the results of a round-robin tournament with 36 competing strategies that was motivated by the work by Axelrod

et al.

on the iterated Prisoner’s Dilemma. We then generalize the “default” version of Iterated TD with respect to two important game parameters, the bonus value and the “granularity” of the allowable bids. We analytically show the impact of the ratio of these two parameters on the game structure. Third, we re-run the 36-player round-robin tournament and investigate how varying the bonus-to-granularity ratio affects relative performances of various types of strategies in the tournament. We draw some conclusions based on those results and outline some promising ways forward in further investigating games whose structures seem to defy the prescriptions of classical game theory.

Predrag T. Tošić, Philip Dasler

Key Distribution Protocol for Secure Multicast with Reduced Communication Delay

Providing effective security with minimum communication cost in the distribution of keying material in a secure multicast communication is a challenging issue, since the amount of information sent through the wired or wireless channel is high. Moreover, the key must be sent securely to the group members. In this paper, we propose a new Key Distribution Protocol that reduces the time delay taken to distribute the keying information to receiver side. To minimize the amount of bits communicated we propose secure one way hash function in the key distribution protocol in which Euler’s Totient Function




) is used in the key computation process. It decrease the time delay to reach the destination by reducing the number of bits communicated while distributing the Keying and re-keying information. Two major operations in this scheme are joining and leaving operations for managing multicast group memberships. The main advantage of this approach is that it uses an existing binary tree based key management scheme to manage the keys that are generated by the GC. The performance of this approach is compared with the existing approaches in this paper and it is shown that this proposed approach takes less communication delay.

P. Vijayakumar, S. Bose, A. Kannan, P. H. Himesh

Special Session on Technology Intelligence

Smart Searching System for Virtual Science Brain

To decide research topics or analyze technical trends, researchers should collect and analyze information from hundreds of thousands of articles, patents, and technical reports. To facilitate the process, information extraction techniques from literature are very helpful. In addition, effective searching methods of the extracted information are necessary as well. While information extraction research has been a popular issue, research about searching and browsing methods for the extracted information has not been an attractive issue relatively. This paper presents a smart searching system that provides various analysis tools, and we expect that researchers can discover and develop new research outcomes through the proposed searching system.

Hong-Woo Chun, Chang-Hoo Jeong, Sa-Kwang Song, Yun-Soo Choi, Do-Heon Jeong, Sung-Pil Choi, Won-Kyung Sung

Using Semantic Web Technologies for Technology Intelligence Services

The change in technological environment presents threats as well as opportunities to companies in the related fields. Existing Technology Intelligence procedures require complicated techniques and high-skilled labor results. Large expert-interviews and manual work is also needed, so single small companies can not undertake this alone. To spread and activate Technology Intelligence in research and industrial fields, we propose shallow, but automated, Technology Intelligence services based on Semantic Web technologies, which can reduce the amount of labor required from experts. We explain our Semantic Web technologies, such as ontology modeling, semantic repository, inference and verification and how they make our Technology Intelligence services possible.

Seungwoo Lee, Mikyoung Lee, Hanmin Jung, Pyung Kim, Dongmin Seo, Tae Hong Kim, Jinhee Lee, Won-Kyung Sung

Procedural Knowledge Extraction on MEDLINE Abstracts

Text mining is a popular methodology for building Technology Intelligence which helps companies or organizations to make better decisions by providing knowledge about the state-of-the-art technologies obtained from the Internet or inside companies. As a matter of fact, the objects or events (so-called declarative knowledge) are the target knowledge that text miners want to catch in general. However, we propose how to extract procedural knowledge rather than declarative knowledge utilizing machine learning method with deep language processing features, as well as how to model it. We show the representation of procedural knowledge in MEDLINE abstracts and provide experiments that are quite promising in that it shows 82% and 63% performances of purpose/solutions (two components of procedural knowledge model) extraction and unit process (basic unit of purpose/solutions) identification respectively, even though we applied strict guidelines in evaluating the performance.

Sa-kwang Song, Heung-seon Oh, Sung Hyon Myaeng, Sung-pil Choi, Hong-woo Chun, Yun-soo Choi, Chang-hoo Jeong


Weitere Informationen

Premium Partner