Catálogo de publicaciones - libros

In the collaborative learning, the learner often focuses on the particular person based on the person’s understanding level. If the information about particular person whom the learner focuses on is acquired automatically, the learner is able to understand the target person easily. So, it is necessary to estimate the understanding levels of others. However, in the collaborative learning environment, since the learner does not utter all the knowledge that he knows, to externalize explicitly the understanding levels of the learners is difficult. The understanding level about the knowledge which is not uttered by the learner is also estimated from the uttered information. So, in this paper, we define solution network which represents the relations of knowledge in the exercise with their strengths. When the utterance is generated, the understanding level of the uttered knowledge and its related knowledge is estimated by means of the solution network.

We have constructed the agent that takes a role of a teacher and assists the group for deriving answer of mathematical exercises in the collaborative learning support system. In mathematical exercises, the complementary diagram is drawn for the purpose of grasping the current learning situation and deducing the derivation method for the next answering step. So, the objective of our research is to construct the advising mechanism for the agent that proposes the diagram in the next answering step. In the environment that students can freely draw a diagram, it is unsuitable to prepare the appropriate diagram for each answering step in advance. So, our system analyzes the students’ diagram and generates a new diagram dynamically based on the analyzed information. In our approach, stepwise generation of the complementary diagram corresponding to each step is regarded as a forward reasoning in the production system. The situation is regarded as the learner’s diagram and the prepared rules are regarded as rules for drawing diagram. So, the diagram for each answering step is generated dynamically by applying particular rules to the current diagram. Consequently, by the proposed diagram, students can arrange the current learning situation and understand the derivation method for the next answering step easily.

Recently, the advent of Internet introduced E-learning, although there are still many practical problems to be solved. This paper presents an educational software framework based on the information technologies called MOLEAS (MOdular Learning Engines with Active Search). The software framework is modularized with standardized interface based on IEEE 1484 LTSA (Learning Technology Standard Architecture) so that the user can reuse other users’ modules to save time and effort. Using peer-to-peer technology, a user can search other users with same educational interests over Internet.

The development of information communication technologies has brought great changes to our society. These advancements enable people to work more efficiently and make daily life more convenient. However, the pervasive application of information communication technologies also has some side effects such as hacking and computer game addiction. While current school education on computers and communication for the gifted students has mainly focused on knowledge and skills on computer, ethical issues on computer knowledge and technologies have not been dealt with sufficiently. In this paper, we introduce a Web-based information communication ethics education system for the gifted elementary school students in computer. The proposed system has the following characteristics: First, extensive contents for the gifted elementary school students are organized. Second, the system adopts various types of avatars for delivering hard contents in an interesting manner. Third, the system supports various types of interactions by providing interactive environments. Finally, the proposed system provides online survey to check students’ progress in information communication ethics.

Web-based learning, be it in higher education, corporate, government or health care sector, has been rapidly developing over the past several years. The current manual system of allocating labs to students is not only time consuming for lecturers and head tutors but is also inefficient and inconvenient. In this paper, we present an evaluation on the design of a web based lab administration system; the system provides a flexible and comprehensive tool for e-learning. The software development life cycle is described, as well as functions which are innovative and unique in this system.

In this paper we describe our experiences with multi-programming (pair and tri programming) in a second level course. The course, “Application Software Development” is a second year course, which has a heavy emphasis on java programming in the labs as well as the development of a full-fledged project. The objective of the course is to build an entire project that comprises of various software engineering activities that span across the semester. In general, we observe that multi-programming improves the students’ ability in analytical thinking and communicating the conceptual ideas. It also raises certain issues when this approach is adopted in the educational context. In this paper we discuss some of these issues. Overall, multi-programming experience has been a rewarding experience for the students in spite of certain problems that were encountered.

We introduce an intelligent safety verification method based on a paraconsistent logic program EVALPSN with a simple example for brewery pipeline valve control. The safety verification is carried out by paraconsistent logic programming called EVALPSN.

Paraconsistent artificial neural network (PANN) is a mathematical structure based on paraconsistent logic, which allows dealing with uncertainties and contradictions. In this paper we propose an application of PANN to analyze cephalometric measurements in order to support orthodontics diagnostics. Orthodontic and cephalometrical analysis taking into account several uncertainties and contradictions, ideal scenario to be treated by paraconsistent approach.

In this work we present some of our investigations concerning paraconsistent, paracomplete and non-alethic systems for knowledge representation in AI. Inconsistency and paracompleteness are natural phenomena in describing parts of our reality. Thus, we need more generical tools, i.e., formal systems for modeling contradictory and paracomplete knowledge, and in this paper we focus non-alethic knowledge in distributed systems.

In this paper, we describe a model for reasoning using forward chaining for predicate logic rules and facts with coarse-coded distributed representations for instantiated predicates in a connectionist frame work. Distributed representations are known to give advantages of good generalization, error correction and graceful degradation of performance under noise conditions. The system supports usage of complex rules which involve multiple conjunctions and disjunctions. The system solves the variable binding problem in a new way using coarse-coded distributed representations of instantiated predicates without the need to decode them into localist representations. Its performance with regard to generalization on unseen inputs and its ability to exhibit fault tolerance under noise conditions is studied and has been found to give good results.