Advances in Web-Based Learning – ICWL 2010

This paper presents a Student Model Repository as well as an IMS LD Authoring Tool that follows the FLEXO approach aimed at supporting adaptive learning using IMS LD. Current IMS LD tools typically manage ad-hoc representations of student models that limit both the scope and possibilities of the adaptation mechanisms. Under the proposed approach, an independent Student Model Repository (1) exposes a set of possible student model attributes that could be used as adaptive variables in any Authoring Tool, and (2) offers to any execution engine, real-time access to the values of those student attributes to allow the execution of adaptive rules embedded in units of learning.

Many personal learning environments (PLEs) are generic collections of available Web 2.0 tools and aggregated information sources. We argue that PLEs are targeting special learning needs of learners in their learning communities. Based on learning content of Chinese Three Hundred Tang Poems, we propose and realize a context-aware multi-granular tagging model to enhance PLEs for language learning.

In this paper, the use of semantic web parsing technique in extracting learning design elements from existing course materials published on the web is described. The rationale of this work is to assist the authoring process of IMS Learning Design (IMS-LD) and thus boosting the adoption of the standard as a publication format of course materials on the Internet. By adopting IMS-LD, learners can be guided to consume the course materials in certain orders decided by course designers. This work is illustrated by implementing a plug-in for the ReCourse Learning Design editor.

This study aims to provide learners with appropriate learning activities navigation based on a learning process recommendation system that gradually adapts to their needs. A learning activity consists of a series of purposeful learning actions with a specific sequence. In the system, user profile is generated from various learning related activities that are collected over different time spans. A dynamic user group is generated based on the similarity of user profiles which have had successful learning experience, and can be used to create a personalized navigation of learning activities for a target learner. In this paper, after introducing how to create a dynamic design of learning activities navigation, we focus on discussing the design and implementation issues of the prototype system for personalized learning support.

In this study, we established an advanced learning environment that aims to promote learning of social communication skills for children especially those with Autism Spectrum Disorder. The learning environment estimates in real time the affective & cognitive state of a child via recognition of multi-modal social signals, and generates interactive narratives with embodied virtual characters. One of the key components of the environment is the Visual Inputs Processor, which is the first that has the capabilities of detecting a child’s attention and expression simultaneously in a natural environment. Furthermore, those have been enabled in a nonintrusive manner which avoids potential bias to the user’s behaviours introduced by intrusive counterparts. The environment also employs multiple inexpensive cameras and a large multi-touch screen, and these settings (1) maximize of space and angles in which observation may be performed and (2) provide users with a much more realistic experience than counterpart approaches do.

Adaptive Educational Hypermedia Systems provide an alternative to the “one-size-fits-all” approach to web-based learning. It aims at enhancing e-learning experience with adaptive capabilities to accommodate for the different learning preference and learning aptitude of individual learner. In this paper, we discuss an Ontology-based Framework for Personalized Adaptive Learning (OPAL). Through the development of OPAL, we implemented an online Java programming course, which provides adaptive features through an ontology model and user model relating to user preference, progress and performance. The evaluation with students shows that the adaptive e-learning framework is more effective for university students who have online learning experience.

Recently, the teaching-learning paradigm is focusing on learners. The individual’s right to select curriculum is gaining ground. As this selection right increases, there is increasing concern and more time needs to be invested in selecting the curriculum suitable for an individual’s situation and preferences. Therefore, an individualized service that can recommend a desirable curriculum to individuals is needed to minimize individuals’ efforts and help them make the right choices. This paper proposes a curriculum recommender system through which individual learners can get advice when they enroll. This research provides the foundation of learner-oriented education by providing a personalized curriculum from the beginning of a course of study.

A lot of research has been done with respect to automated evaluation of students’ knowledge. Learning outcomes have been of serious interest to many research communities as well. In this article we apply dominance relations in rough sets approximations for assessing knowledge obtained in active learning environments.

This paper presents the results of our efforts to incorporate automated judging into the evaluation of programming assignments by students. Our experimentation on automated evaluation of programming assignments was twofold: at first, we have used the stand alone open source system Mooshak independently from the learning management system supporting the rest of e-learning activities; then, we have isolated and integrated its automatic judge engine into the learning management system Claroline. In both cases, automated judging has been used in conjunction with marking by instructors offering regular and frequent feedback. An online questionnaire was presented to the students to quantify their preferences in using the automatic judge with respect to pedagogical results. The data we gathered provides a clear indication that students think positively of the automatic judge when used for evaluation of programming assignments and in conjunction with their instructors’ feedback and marking.

With the large uptake of online learning environments in educational institutions, students and teachers interact no longer solely face-to-face and are often geographically dispersed. Distance education can make it much harder for students to position themselves in the group of fellow students and to self-reflect on their work. The teacher can have similar problems by losing the general overview of the class, which makes it hard to discover potential pitfalls of students in time. To assist with this problem, we have developed a set of visualizations of learner activities to increase awareness and to support self-reflection. This paper discusses those visualizations and evaluates the usability and user satisfaction. From these studies, first preliminary results are available of the usefulness of the tool for the students. A planned future evaluation is discussed to evaluate the real usage and impact of the tool.

Learning Style Index is a method to classify the learning preferences for learners. Learning preferences can then help learners to find their better way in the learning process. It also can help teachers to adopt suitable learning materials for efficient learning. This paper is concerned on the study, implementation, and application of a web-based learning style index. We also show a case study on the integration of learning style index into an e-learning system.

Content development for adaptive educational systems is known to be an error-prone task. Gaps can occur when the content is created, modified or when the context of its usage changes. This paper aims at improving the existing practises of learning content quality control in adaptive educational systems by automating the detection and management of gaps. Several categories of gaps are identified to account for structural, linguistic, and semantic inconsistencies in collections of learning content. An effective filtering mechanism has been implemented in order to separate the collected gap data into categories that are relevant for the current authoring context. An evaluation of the developed tool demonstrates its utility.

We describe a web-based system that helps undergraduate students learn about research at their university through the medium of textbooks. The system, called University on Textbooks, links textbooks to research articles written by university researchers, enabling people to navigate between introductory materials in textbooks and specific topics reported in research articles. Bidirectional links are generated from semantic statements created manually for topical passages from textbooks and abstracts of articles. The semantic statements are created using an ontology that is based on a description logic, so a reasoner can infer the semantic similarity between pairs of statements. To demonstrate the feasibility and effectiveness of our approach, we present a prototype containing 100 semantic statements describing passages from an undergraduate life sciences textbook in Japanese that have been linked to semantic statements created for 392 research articles from life sciences.

This paper presents Oscar, a conversational intelligent tutoring system (CITS) which dynamically predicts and adapts to a student’s learning style throughout the tutoring conversation. Oscar aims to mimic a human tutor to improve the effectiveness of the learning experience by leading a natural language tutorial and modifying the tutoring style to suit an individual’s learning style. Intelligent solution analysis and support have been incorporated to help students establish a deeper understanding of the topic and boost confidence. Oscar CITS with its natural dialogue interface and classroom tutorial style is more intuitive to learners than learning systems designed specifically to capture learning styles. An initial study is reported which produced encouraging results in predicting several learning styles and positive test score improvements in all students across the sample.

The e-Framework is arguably the most prominent e-learning framework currently in use. For this reason it was selected as basis for modelling a programming exercises evaluation service. The purpose of this type of evaluator is to mark and grade exercises in computer programming courses and in programming contests. By exposing its functions as services a programming exercise evaluator is able to participate in business processes integrating different system types, such as Programming Contest Management Systems, Learning Management Systems, Integrated Development Environments and Learning Object Repositories. This paper formalizes the approaches to be used in the implementation of a programming exercise evaluator as a service on the e-Framework.

With the fast growing of using herbal medicine, pharmacists now need a basic knowledge of these topics for their professional practices. To serve the need, a set of courses on herbal medicine, is arranged for pharmacy students. Due to a limitation of time, it is hard for a student to familiar with medicinal herbs. In this paper, we introduce KUIHerbRx, a Web-based supplement learning tool on herbal medicine. The KUIHerbRx is a modified version of the Knowledge Unifying Initiator for Herbal Information (KUIHerb), which is used as a platform for building a Web community for collecting the intercultural herbal knowledge with the concept of a collective intelligence. Due to the diversities of herbs, geographic distribution and their applications, social network collaboration is important for enhanced learning. Three types of information creation, i.e., initial, voting and non-voting information, is designed. KUIHerbRx provides a supplement learning to improve knowledge and skill in herbal medicine with a scientific method. Information of medicinal herbs in several regions can be distributed and exchanged among groups of students.

This paper discusses the potential role of social software in supporting teamwork and collaborative project management in higher education. Based on the fact that social software has been widely spread among young students nowadays, using it for collaborative learning is believed to increase students’ involvement and create learning incentives. Two social software platforms,

Graaasp

and

Google Wave

are examined in terms of sustaining collaborative learning activities. Relevant existing features and possible extensions that enhance the learning experience are addressed. Benefits and challenges resulting from the bottom-up learning paradigm are also presented.

Distributed cognitive learning theory provides a comprehensive model to cognitive interaction and knowledge sharing. If distributed cognitive theory and distributed virtual reality technology are introduced into intelligent tutoring system, the creation of a realistic cognitive context, the sharing of resources, the improvement of personalized learning and the enhancement of teaching efficiency will become possible. This paper discusses distributed cognitive learning theory and its guidance to system design, presents the DITS (Distributed Intelligent Tutoring System, DITS) architecture based on Multi-Agent System (MAS) and distributed cognitive theory, and focuses on the analysis of the learning services core modules of DITS, including organization and management of teaching, learning, teaching experts and collaborative communication.

This research provides students a three-phased English essay reading scaffolding support system by leading them through a systematic process. By integrating Reading annotator, a web-based annotation system for reading support, the system can help students know how to enhance their reading abilities. Reading annotator helps students visual interactively to analyze basic structure of essays by identifying key elements of each paragraph such as topic sentence, controlling idea, and supporting details. Based on the key elements of each paragraph, a process-oriented structure-based scaffolding support environment is developed to help students know how to read essays with various rhetorical forms step-by-step by analyzing the essay organizations. The system applies not only text structure, visual representations, summarization, and questions reading strategies but also reading to summarize and reading to analyze reading levels. Moreover, the proposed system accommodates the scaffolding strategies such as providing procedural facilitator, guided learning, using external representational scaffolds, helping the student focus thinking on only those items that were important for the task.

The ubiquitous availability of mobile communication devices which are connected to the Internet makes it possible to dedicate small amounts of spare time for mobile learning. The limited screen size and resolution makes current mobile devices quite hard to visualize multifaceted Web pages, so this kind of content could be adjusted to meet the device needs. Adjustment in e-Learning means the contextualization of learning contents for m-Learning usage. In this paper, we present a design and implementation of a Web service-based system for adapting, displaying and manipulating learning objects on small handheld devices. A speech solution allows learners to turn any written text into natural speech files, in using standard voices. Our approach allows generation of text and audio based learning material dynamically for m-Learning and ubiquitous access. The deployment of audio data to the Web user via progressive download is also presented.

In CLIL (Content and Language Integrated Learning) the context is to explicitly teach a subject through a foreign or second language but also implicitly to teach a language through a subject; adding the activity of social networking is part of making the context, and hence language learning, real. In this paper we describe the learning design and infrastructure of a CLIL seminar for trainee teachers. Active learning is supported by our LearnWeb2.0 infrastructure which helps students in collaborative searching and organization of resources to design CLIL materials for their future pupils. In LearnWeb2.0 searching is focused on a specific subject providing materials from appropriate Web2.0 platforms, ratings and comments are directly linked to the resources and easily shared within the learning group. We also present our evaluation design, as well as the first promising evaluation results.

Since Bloom’s initial work on competencies in 1956, various competency systems have been designed and used to assess students’ competencies. Different pedagogical researchers and stakeholders prefer different systems. We have been collaborating with them. Such systems are essential for the adaptation by adaptive intelligent tutoring systems. Now, this paper presents how

ActiveMath

integrates several competency systems to bridge the gap between different competency systems and thereby facilitating the reuse of learning objects across system boundaries. The combination of competency-related data is achieved by mapping a new competency system to the internal one.

Learning objects have been the promise of providing people with high quality learning resources. Initiatives such as MIT OpenCourseWare, MERLOT and others have shown the real possibilities of creating and sharing knowledge through Internet. Thousands of educational resources are available through learning object repositories. We indeed live in an age of content abundance, and content can be considered as infrastructure for building adaptive and personalized learning paths, promoting both formal and informal learning. Nevertheless, although most educational institutions are adopting a more open approach, publishing huge amounts of educational resources, the reality is that these resources are barely used in other educational contexts. This paradox can be partly explained by the difficulties in adapting such resources with respect to language, e-learning standards and specifications and, finally, granularity. Furthermore, if we want our learners to use and take advantage of learning object repositories, we need to provide them with additional services than just browsing and searching for resources. Social networks can be a first step towards creating an open social community of learning around a topic or a subject. In this paper we discuss and analyze the process of using a learning object repository and building a social network on the top of it, with respect to the information architecture needed to capture and store the interaction between learners and resources in form of learning object metadata

Hybrid Learning is becoming one of important applications by integrating e-learning and traditional face-to-face instruction together. As an introduction to the reader the challenges in hybrid learning of physical education and architecture of Physical Education Hybrid Learning Platform (PEHLP) are firstly provided. The main character of PEHLP is using video instead of textual information as learning material. Then Video Review Module (VRM) is presented. Through the annotation composing functionality of VRM, the teacher can review the student’s actions video, pick out the wrong actions and add still annotations, which for now are either text or specific graphic symbols or audio-recording. And through the annotation viewing/browsing functionality, the student can watch annotated video, find out mistakes and get instructions. In order to control over bandwidths and promote video annotation, the geometry primitives of video annotation graphic symbols are presented in Geometry Primitive Language (GPL) in XML. The experiment shows that the annotation data is 0.37% of the annotation data in JPG format. The user case of this paper shows that 80.9% out of students think that the video review can afford the hybrid learning of physical education, but only 66.7% of them think the hybrid learning instruction mode can attain the same teaching effect in contrasted F2F instruction mode. So we suggest that the proportion of e-learning should be less than 60% in hybrid learning of physical education course.

Emotions play an important role in an e-learning process. Optimal emotions can enhance learning enthusiasm, but negative emotions can suppress learning’s interest. According to the basic emotion regulation theory, the paper proposes a model of emotion regulation, and provides an optimal mood to learners in an e-Learning environment. The model evaluates an e-learner’s current emotions with emotional lexicon and semantic rules, and uses the emotion regulation strategy scale to determine the learner’s strategy of emotional regulation, then uses specific adjustments to remedy negative emotions presented to the e-learner. Finally, the paper presents a prototype that has been built to validate the model in an emotion-chat system.

Although online tests are a quite popular form of assessing learning outcomes in traditional Learning Management Systems, it is quite difficult to deliver the wealth of existing re-usable QTI-compliant tests into Web 2.0 style Personal Learning Environments. This paper addresses both pedagogical scenarios and technical solution for using QTI self-tests in blogs, personal e-portfolios and social media platforms. The empirical part describes the development of the original QTI Assessment Proxy Service and QTI self-test player based on the Wookie widget engine.

This paper presents

e-Training DS

, an instructor-oriented toolkit for the integration of portable game consoles in modern e-learning environments (e.g.

Moodle

TM

). The tool is based on a general model in which the instructor maintains a library of mini-games that are easily produced by customizing generic game patterns (e.g. a simple jigsaw puzzle). The authors can then compile a working set of mini-games and distribute it to the students so that they can play using their own portable game devices. Moreover, the game packs produced with

e-Training DS

can track the students’ activity and send the results to the back-end e-learning environment. In this manner instructors can evaluate the students’ performance using the facilities provided by the e-learning system. While

e-Training DS

is focused on the Computer Science field and the

Nintendo DS

TM

game console, the approach proposed can be repurposed for other knowledge areas and devices.

A Personal Learning Environment (PLE) is a mash-up of learning services. It enables students and teachers to assemble a work environment that is adapted to a domain and specific individual needs. In this article, we report on our experiences on using a PLE for Language Learning in five French lectures at the Shanghai Jiao Tong University Continuing Education School. We found that while a PLE has the potential to simplify access to and usage of Web sites and services for language learning, students will use it only if properly motivated. Furthermore, at the time being, difficulties that result from the user interface and technical implementation make the interactions with PLEs difficult. The problems need to be overcome in order for PLEs to become adopted by the average, not technically highly literate students and teachers.

Many tools have been developed as high-level tools of Learning Design. However, few of them concern about the teacher’s perspective. This paper proposes a platform for generating online course structure adapted to teachers’ profile. Based on quantitative research, teacher’s features are observed and integrated into Unit of Learning models, and these models are presented in the platform e-Course Help, using a presentation of terms and structures distant from the IMS LD specification.

The field of collaborative learning has been researched for many years with a focus on evaluation of the circumstances under which collaborative learning provides better learning effectiveness compared to individual learning. Traditional collaborative learning has been taking place as group work or discussions in a class. Recently, Serious Games for learning and Digital Educational Games (DEGs) have been established as a promising alternative to the traditional one-to-many teaching as it is still standard in today’s classrooms and with the uprising technology of Multiplayer Online Games (MOGs), new ways of collaborative learning are possible. In this paper we examine various chances and challenges for collaborative learning which arise by the use of Multiplayer Serious Games. Section[1] motivates this work, followed by a description of some important aspects of MOGs and collaboration in Section[2]. In Section[3] the concept of collaborative learning is explained and important concepts and methods for a collaborative game design are analyzed and discussed in Section[4], using some examples of collaborative Serious Games. Finally, we briefly introduce our own approach of a collaborative Serious MOG in Section[5]. In Section[6] the results are summarized and further work is pointed out.

This paper describes and evaluates CONSPECT, a service that analyses states in a learner’s conceptual development. CONSPECT combines two technologies – Latent Semantic Analysis (LSA) and Network Analysis (NA) into a technique called Meaningful Interaction Analysis (MIA). It uses LSA for the language analysis and NA to provide visualisations of the semantic relatedness information calculated by LSA. CONSPECT was designed to help both learners and tutors monitor conceptual development. This paper reports on the verification activities undertaken to show how well LSA matches humans in clustering similar concepts. The verification used the edit distance between card sorts to quantitatively evaluate the service.

This paper proposes an agent-based virtual teaching assistant, called LABTA. It can be applied to various experiment courses. LABTA provides students with the basic functionality for experiment courses: guiding, demonstrating, and explaining. LABTA supports both keyword and sentence based query and possesses an adaptive learning ability to gain knowledge from other agents or knowledge sources. The experimental results showed that LABTA is efficient to guide and assist students in solving their experimental problems.

Learning path construction is a complex task. It involves formulating and organizing learning activities, defining ways to evaluate student learning progress and to match such progress with designated learning outcome requirements. Existing methods for adaptive learning path construction typically consider knowledge elements (KEs) as the building blocks and offer mechanisms to select and schedule relevant KEs to form a learning path. However, as KEs may involve variety of delivery and assessment methods, relying on KEs to form building blocks can make learning path construction difficult. To avoid this problem, existing methods restrict KEs to be delivered only through “lecturing” and “question-answering” activities, even though such a restriction greatly affects the usefulness of these methods. In this paper, we propose an open model to formulate learning paths and learning activities. This model can lead to the implementation of a generic system to support learning path design for teachers from any subject disciplines. We have developed a prototype based on this model and conducted a user study to evaluate its effectiveness.

In recent years, E-learning has become a popular research topic with its wide applications. As an important factor of E-learning, concept map is a powerful way to manage knowledge. In this paper, we describe our approach for determining a compact concept map of an input dance motion obtained from the 3D motion capture technology. The dance motion is analyzed to extract the repetitive patterns and then the prerequisite relations are computed as the inclusion relation among patterns. A concept map can then be constructed automatically for illustrating such relations. The transitive reduction algorithm is applied to remove the redundant edges such that the resulting concept map retains a compact representation of the dance structure. This concept map can be used to generate dance lessons such that a learner can learn the dance motion in a step-by-step and logical manner.

Game-based learning or serious games is becoming an important trend in the e-learning research area and seems address several typical e-learning problems such as high dropout rates, due to the lack of motivation to continue studying. In serious games, it is very hard to define and mix the learning situations with the game characteristics, and to integrate an assessment and guidance process of the learner without disturbing the game progress and maintain the intrinsic characteristics of the video game: fun, player motivation, immersion and interaction. In this paper, we consider the serious game as an asynchronous and concurrent system, and we propose an approach based on a Petri net to assess learners and detect misconceptions. In the game design stage, a discussion between domain experts, learning experts, and game designers is engaged in order to identify the actions in the game that imply knowledge acquisition and allow achieving the learning objectives of levels. Therefore, in our approach, the Petri net models only game actions allowing the learner to acquire knowledge. We use the reachability graph of the Petri net to track the learner in order to detect, in real time, the learner’s misconceptions, improve learner assessment and provide an accurate feed back for both the learner and the instructor.

This paper proposes an efficient probabilistic indexing scheme called

P

robabilistic

R

etrieval

-

Tree

(PR-Tree) to facilitate a efficient probabilistic retrieval of Chinese calligraphic manuscript images based on triple features such as contour points, character styles and number of strokes. To the best of our understanding, this is first work on probabilistic retrieval over Chinese character. Different from conventional character retrieval and indexing methods [18] which only adopts shape similarity as a query metric, our proposed indexing algorithm allows user to choose the above three kinds of features as query elements. Moreover, a probabilistic model is introduced into the character retrieval process. Comprehensive experiments are conducted to testify the effectiveness and efficiency of our proposed retrieval and indexing methods respectively.

Building a good service environment for the participants is always the whole goal of smart classroom, as well as a too complex goal to achieve in several projects, due to involving various aspects. In this paper we explore a way approaching the goal by providing the service in an attentive mode effectively. For this goal, we develop some underlying tools, and integrate those tools into the system of Smart Classroom. By this means, in some typical scenarios, the smart classroom can be aware of the situation and the teacher’s intention automatically and then provide the relevant service or assistance actively to meet the participants’ need or improve the work efficiency.

The purpose of this study is to see whether students in experimental class outscore their counterparts in control class in writing and what students’ attitudes towards cooperative learning are. The experiment lasts 17 weeks. The main instruments in the experiment are pre-test, post-test and questionnaires. The results show that: (1) There are no significant differences between control class and experimental class in writing scores before and after the experiment; and (2) the students have a positive attitude towards cooperative learning in English writing. Most of them become more confident, like learning from each other and have better writing habits.

Appropriately designed web based learning materials are thought to be more efficient than traditional ones because of interactivity and multimedia, which enable clear explanations, motivating design, active learning, instant feedback and other advantages. Our research question is whether students achieve better learning results if they learn from web based learning materials than from traditional ones. Although the research results show that students spent less time learning from traditional learning materials than students learning from web based learning materials, the differences of learning times were not significant. However, we proved significant differences in terms of gained knowledge. The students using web based learning materials performed significantly better than students using traditional learning materials.

Computerized adaptive testing (CAT) is a method of administering tests that adapts to the examinee’s ability level. Previous research has focused on estimating the examinee’s ability accurately and on providing adequate feedback upon analyzing the examinee’s ability. However, in order for students to use the feedback, they must find courses or learning materials themselves. It is difficult to make customized learning available continuously. Therefore, we used adaptive testing to estimate a student’s ability and to identify a number of student characteristics. This paper recommends content that can reinforce areas in which the student needs improvement. We applied our system at an actual education site. The group that used our recommendation module learned more effectively than the control group. By using this system, teachers will be able to monitor students closely. This enables customized learning which allows students to study effectively without necessitating the effort to search for learning materials. Customized learning will increase interest in.

In this study, we develop an innovative approach using sharable learning activity to create an open learning network. In this paper, we present our vision of open learning and a framework for sharable learning activities. We further discuss how to utilize open source software to implement a prototype system, and describe the scenarios and features of the proposed framework and system.

Compared with traditional teaching service, Web-based learning system is greatly different in many aspects like business model, participants, consumers, technique facilities and etc.; therefore, studying design strategies for Web-based learning system has great significance. The forming process of design strategies for a Web-based English textbook and teaching service project is introduced in this thesis. From the perspective of sustainable design, this article analyzes demands of investors, participants, and consumers thoroughly. On the basis of analysis, originalities of design strategy are created by taking advantage of features of Web technique. This paper also presents design ideas of constructing a product-service system of English teaching, realizing distribution according to functions, realizing self-update of textbook.

In virtual experiment environment, it is very important that users can collaborate to do an experiment. Based on the combination of collaboration technology and virtual experiment, the paper has proposed a collaboration support platform. In the platform, all users who will do collaboration form a group, in which there exists a master user. In the process of collaboration, the master user creates packets which includes operations of the master user and sends to other users through the application server. The other users can restore the operations in their terminal. The application server ensures the synchronization of collaboration. According to the analysis of running instance, the collaboration system has dynamic adaptability and distribution of collaborative capabilities and efficiently solves the problem of resource sharing and scheduling in the virtual experiment environment.

This paper describes a pedagogical pattern for cross-cultural language learning and its application in a case study involving learners in Australia and China. The pattern uses a multimedia discussion tool as its main component, supported by video conferencing and chat. We analyze how students interacted with the tools, especially the documents they created and the commenting behavior. Based on this analysis, we give a set of recommendations relevant to practitioners and researchers in technology-enhanced language learning.