Catálogo de publicaciones - libros

Web-based learning has the potential to becoming an indispensable part of learning by augmenting traditional in-class, teacher-led education. Although there have been several noteworthy signs of advances and adoption of Web-based learning, there is a long way to go before the full potential of Web-based learning can be realized. In this paper, eight specific directions are given for overcoming the current challenges for Web-based learning.

E-learning systems , as an education pattern, are becoming more and more popular. In e-learning systems, courseware management is an indispensable part. As the number of various courseware increases, how to find the courseware or learning materials that are most suitable to users and users of e-learning systems are most interested in is a practical problem. In this paper, we apply the idea of knowledge discovery techniques to make personalized recommendation for courseware. We design the courseware recommendation algorithm which combines contents filtering and collaborative filtering techniques. Also we propose the architecture of courseware management system with courseware recommendation, which is seamlessly integrated in our E-learning system. The experiment shows that our algorithm is able to truly reflect users’ interests with high efficiency.

With the ability to connect people and information around the world, the Internet is already having a significant impact on the traditional education. Nowadays, students can easily access the online course materials anytime anywhere. As a coin has two sides, Internet also amplifies the complexity of the course materials development and their provisioning. Since the learning initiative is taken by a student in the e-learning environment, traditional teacher-driven learning model is no longer applicable. Thus, course materials which are prepared based on individual student’s and become critical. In this paper, we introduce a model for personalized course material generation through 1) student ability test based on Item Response Theory, 2) a Dynamic Conceptual Network for course materials management, and 3) intelligent knowledge base facilities, such as personal profiles, to help understand students’ behaviors so as to materialize the concept of personalization.

Thinking is as important as learning. It is one of important and effective learning approaches by recording and reflecting student’s own learning attitude and learning performance and referring to others’ learning portfolio daily. Although many of students know this approach, but seldom of them put this approach into action owing to the problem of preserving content of reflection and learning portfolio. In this paper, a networked learning reflection mechanism based on blog service is presented. Students can make their learning reflection and view others’ with ease. An experiment was conducted at an Institute of Technology upon system completion. The experimental results show that there is a positive correlation between students’ learning performance and attitude of learning reflection. It reveals that students who have better learning performance tend to make more reflection on their learning. More than 80% of students express this mechanism has positive affection on their learning.

Personalized recommendation service based on the web is the current research focus of intelligent information retrieval and distant education. In this paper, an intelligent recommendation system is proposed. The system depends on the mining results of Web log and cache data, comprehensively evaluates the influence of navigation times, browsing period, and page volume. Through effective classification of the interested pages from Web log and cache data, the interest model of different user groups is established. Meanwhile, a method for recommending the valuable knowledge is put forward, which combines the approach of content-based filtering and collaborative filtering. Experiments demonstrate that the proposed recommendation method is feasible and effective.

E-learning through the web began as a service, by publishing some educational materials at websites. At the early stages, the learning process was carried out by browsing such a collection of educational materials and its effectiveness was based on the of content in those materials. Customizing the presentation and interactivity of an e-learning application based on the users (which we call here personalization), should be done based on each learner’s knowledge and learning experience. A learning pattern, mechanism presented in this paper, is a meta-level piece of information that can be used to achieve low coupling personalization in an e-learning courseware. A learning pattern, which is modeled on different levels of abstraction, is constructed based on the interaction of a learner with relevant materials and it is depreciated when he/she stops accessing such materials. In this paper, we also present how a learning pattern supports to define domain knowledge in SCORM compatible environment.

A method for capturing the interest and authority of students about course content is proposed and implemented as a user modeling approach in a Web-based user-interactive question-answering (QA) system. An instructor has to define a topic ontology (or concept hierarchy) for the course content so that the system can generate the corresponding structure of boards to hold relevant questions. The students can interactively post questions, and browse, select, and answer others’ questions in their interested boards. The users’ log data are accumulated and organized as the users’ historical data, which are used to build the association space containing the association relations between the users’ historical data and the topic ontology. From the association space, the interest and authority of students about the questions in each board can be computed first and the interest and authority of students about each topic in the ontology can be computed based on the corresponding parameters of its offspring (sub-topics or questions). These user models (interest and authority) can be used to automatically and properly distribute relevant questions and answers to relevant students to enhance learning efficiency and help instructors design suitable teaching materials to enhance instruction efficiency.

The emergence of Learning Objects in learning technology has had a big impact on the learning environment and its community. Teachers and students have more facilities for using the E-Learning system, which has made the learning process more effective and beneficial. This paper presents an application framework for using Learning Objects in the learning styles application of an E-Learning system for an Object-Oriented C++ Programming subject. The project’s end-user is a higher learning institution student or anyone who is learning the subject. The focus is on the reusability aspect of the note—either content or database. The framework is explained by using SCORM-based architecture design, context diagram, database design, and interface design.

We propose a novel data distribution framework for developing a large Web-based course forum system. In the distributed architectural design, each forum server is full equipped with ability to support some course forums independently. The forum servers collaborating with each other constitute the whole forum system. Therefore the workload of the course forums can be shared by a group of the servers. With the secure group communication protocol and fault tolerance design, the new distribution framework provides a robust and high performance distributed architecture for the large course forum system. The forum servers can be settled in anywhere as long as a broadband network connection to Internet is provided. Our experimental performance testing results show that the large forum system is a high performance distributed system with low network overhead cost. In addition, a semantics based clustering algorithm is developed to classify the courses by the content relevance of their teaching material. Relevant course forums can be arranged on the same forum server together. Hence our distribution framework also provides a knowledge-based taxonomic storage solution to build a large course teaching material digital library.

Although the interactive web-based learning environment enriches the study of lighting engineering, it relies healthy on realistic rendering and illumination calculations of 3D scenes. These operations are typically very expensive in kernel computations and content distribution. Hence, they are not applicable to dynamic scene. Our objective in this work is to address this problem through a multi-resolution approach in order to provide a real-time feedback to the students as they modify the lighting conditions or the geometry of the scene. We apply progressive and selective data transfer of 3D models with geometry and illumination attributes. The experimental results show that the proposed solution provides better user experience to the students.