Catálogo de publicaciones - libros

Research on e-learner community building has attracted much attention for its effectiveness in sharing the learning experience and resources among geographically dispersed e-learners. While collaborative filtering proves its success as one of the most efficient methods in finding similar users in e-commerce domain, it does meet special challenges in e-learning areas. In this paper, we incorporate multi-agent techniques into collaborative filtering and propose a novel community building scheme. By doing so, we manage to collect useful information from the learner behaviors and thus increase the scalability and flexibility of traditional collaborative filtering methods. The experiment on a standard benchmark shows that our scheme has reasonable community building quality and e-learners can make better recommendations to each other inside the community.

For overcoming the existing problems in Web based learning system when network blocking or disconnection happens, a new architecture of the Web based learning system based on semantic caching is proposed; also, the model and the query processing mechanism of semantic caching are described in detail. Furthermore, this paper proposed and proved 11 simplification rules, which can be used to optimize the query processing of semantic caching, and the optimized query-processing algorithm is thoroughly described.

In this paper, we focus on the semantic relationships acquisition from Chinese web documents motivated by the large requirement of web question answering system in e-Learning. With our scheme, we dwindle in numbers of text to be analyzed and obtain initial sentence-level text in pre-process phase. Then linguistic rules, which are broken down into unambiguous and ambiguous, designed for Chinese phrases are applied to these sentence-level text to extract the synonymy relationship, hyponymy relationship, hypernymy relationship and parataxis relationship. Lastly, candidates are refined using two heuristics. Compared to other previous works, we apply not only strict unambiguous linguistic rules but also loose ambiguous linguistic rules to extract relationships and proposed efficient approach to refine the outputs of these rules. Experiments show that this method can acquire semantic relationships efficiently and effectively.

In this paper we investigate the use of Latent Semantic Analysis (LSA), Critiquing Systems, and Knowledge Building to support computer-based teaching of English composition. We have built and tested an English Composition Critiquing System that make use of LSA to analyze student essays and compute feedback by comparing their essays with teacher’s model essays. LSA values are input to a critiquing component to provide a user interface for the students. A software agent can also use the critic feedback to coordinate a collaborative knowledge building session with multiple users (students and teachers). Shared feedback provides seed questions that can trigger discussion and extended reflection about the next phase of writing. We present the first version of a prototype we have built, and report the results from an informal experiment. We end the paper by describing our plans for future work.

Web-based training offers many benefits over instructor-led training environments. It provides a time, class size and geographical location independent learning platform to students. To enable active learning and enhance the effectiveness in students’ understanding of the training materials, multimedia cues, like 3D graphics, animation and sound, have been employed in web-based training systems to achieve these goals. However, if a training system involves a large amount of 3D animation, such as crowd animation in an emergency evacuation training system, the requirements for rendering capability and network bandwidth may become too high to meet. In this paper, we propose an adaptive human motion animation method to support real-time rendering and transmission of human motions in web-based training systems. Our method offers a mechanism to extract human motion data at various levels of detail (LoD). We also propose a set of importance factors to allow a web-based training system to determine the LoD of the human motion for rendering as well as the LoD for transmission, according to the importance of the motion and the available network bandwidth, respectively. We demonstrate the effectiveness of the new method with some experimental results.

Practical exercises and assignments are an integral part of programming course, which is a core basic skill required in computer science and best learned by doing. For effective learning, assignments have to be inspected and graded carefully but quickly so that students may benefit from the useful and timely feedback. This can be facilitated by automating the process using a specifically designed software system. PASS is such a system being developed in our department. A first version has been pilot run, with very encouraging responses from tutors and students. We observed that with this automated system, students can have their programs tested anytime, anywhere. They are encouraged to practice more without hesitation or embarrassment about their programming mistakes, and thus this learning environment can boost up their motivation to continue the practice of their programming skills. This paper aims at sharing the initial experiences we gained in using this system.

In this paper, we propose an automatic leveling system for e-learning examination pool using the algorithm of the decision tree. The automatic leveling system is built to automatically level each question in the examination pool according its difficulty. Thus, an e-learning system can choose questions that are suitable for each learner according to individual background. Not all attributes are relevant to the classification, in other words, the decision tree tells the importance of each attribute.

Learning computer programming is not an easy task. Students need to spend hours doing practical activities in order to comprehend the techniques of writing computer programs and beginners usually face a number of obstacles associated with installing and using a compiler or integrated development environment. This paper introduces an online web-based system that provides an interactive integrated environment for students doing programming activities and coursework in a distance learning institution. The interactive system provides students with timely and effective feedback about programming activities without the need to have instructors and students meet at the same time and the same place. The web-based system provides students with an editing, compiling, testing and debugging environment for learning computer programming on the web. Instructors can monitor the learning progress of students, compile the student’s program and view the error messages through the student’s workplace in the online system.

This paper proposed the design and implementation of digital signal processing virtual lab (DSPVL) based on components. In the DSPVL, all the virtual instruments are developed as components and implemented as Java Beans, which improve the developing efficiency, the reuse of software and make the system be maintained and expanded easily. This paper also introduces the characters and architecture of user platform and illustrates the key design and implementation technologies. In the DSPVL, we developed a lot of components for the experiments in Digital Signal Processing (DSP) course. In this paper, we also gives an example of designing components of Discrete Fourier Transform (DFT) to show the process of designing, implementing and using the components in the DSPVL.

As the learning paradigm shifts to a more personalised learning process, users need dynamic feedback from their knowledge path. Learning Management Systems (LMS) offer customised feedback dependent on questions and the answers given. However these LMSs are not designed to generate personalised feedback for an individual learner, tutor and instructional designer. This paper presents an approach for generating constructive feedback for all stakeholders during a personalised learning process.

The dynamic personalised feedback model generates feedback based on the learning objectives for the Learning Object. Feedback can be generated at Learning Object level and the Information Object level for both the individual learner and the group. The group feedback is meant for the tutors and instructional designer to improve the learning process.