International Journal on Artificial Intelligence Tools

The paper illustrates how fuzzy techniques can be applied to address problems experienced in widespread computing systems for distance learning. Many educational organizations use Web Course Management Systems (WCMS) to run distance courses. In such environments, facilitators often face difficulties in monitoring and understanding problems experienced by distance learners. An approach is proposed here where artificial advisors are built to offer distance-learning facilitators informed advice of what the problems/needs of individuals and groups may be, as well as to suggest appropriate actions when possible. We have developed the TADV (Teacher ADVisor) framework, which builds student, group, and class models based on the tracking information generated by WCMS, and uses these models to generate advice to the course instructors. This paper introduces TADV and describes the fuzzy approach used for extracting individual, group and class models. TADV is currently exemplified in a Discrete Mathematics course run at the Arab Academy for Science and Technology and Maritime Transport, Alexandria, Egypt.

This paper discussed the use of Conceptual Graphs (CGs) for implementing tasks employed in web-based educational systems that aid terminology learning. Specifically, we focus on two critical issues in intelligent tutoring - student diagnosis and generation of adaptive explanations. Both tasks are demonstrated in terminological domains where learners have to familiarize themselves with concepts in a specific subject area (e.g. computing, finance, chemistry). Based on CG reasoning, robust and computationally tractable algorithms for student modelling and adaptive explanation generation are defined. Two intelligent systems are presented — STyLE-OLM and HYLITE+. STyLE-OLM is an interactive learner modelling system that extracts extended models of the learners' cognition. HYLITE+ is a natural language generation system that generates adaptive Web pages based on a learner model(LM). The two systems are complementary and have been implemented separately. However, considered together they cover most of the key tasks in adaptive web-based educational hypermedia that aid learning technical terminology. Based on evaluative studies of STyLE-OLM and HYLITE+, the use of CGs for interactive open student modelling and adaptive concept explanations is examined. The applicability of CGs in adaptive web-based systems that aid learning technical terminology is discussed.

This paper introduces a statistical framework for extracting medical domain knowledge from heterogeneous corpora. The acquired information is incorporated into a natural language understanding agent and applied to DIKTIS, an existing web-based educational dialogue system for the chemotherapy of nosocomial and community acquired pneumonia, aiming at providing a more intelligent natural language interaction. Unlike the majority of existing dialogue understanding engines, the presented system automatically encodes semantic representation of a user's query using Bayesian networks. The structure of the networks is determined from annotated dialogue corpora using the Bayesian scoring method, thus eliminating the tedious and costly process of manually coding domain knowledge. The conditional probability distributions are estimated during a training phase using data obtained from the same set of dialogue acts. In order to cope with words absent from our restricted dialogue corpus, a separate offline module was incorporated, which estimates their semantic role from both medical and general raw text corpora, correlating them with known lexical-semantically similar words or predefined topics. Lexical similarity is identified on the basis of both contextual similarity and co-occurrence in conjunctive expressions. The evaluation of the platform was performed against the existing language natural understanding module of DIKTIS, the architecture of which is based on manually embedded domain knowledge.

This article presents a system that maps classroom lectures directly into web based education lessons. In the classroom, the lecturer writes on a wide, pen sensitive display. The system tracks all actions and makes it possible for the lecture to be replayed from the web any time. The remote viewer can follow the progress of the lecture: Audio, the creation of the board content, and an optional video image of the instructor is transmitted. In addition to usual drawing functionality the board can handle a range of multimedia elements from the Internet. The board can integrate different kinds of modules, invoked by board drawings. One of these modules is described here: A computer algebra system that evaluates mathematical expressions or plots functions is placed at the lecturers disposal by a handwriting recognition.

Nowadays, web-based distance learning is benefiting from improved communication services. However, the mere fact of setting up an environment for students and lecturers does not guarantee mutual collaboration or successful student learning. This is partly due to the fact that an unique response is given to every user with different background knowledge, different interests, or different skill levels in the use of the services provided. To resolve this situation, adaptive systems provide an adapted response to each user's needs based on a constructed user model containing his/her characteristics. In this paper we will see that the construction of this user model is not trivial. In a user model in a web-based collaborative environment, there are very diverse attributes related with the interaction of the user with the services and the value of these attributes are obtained in very different ways. In addition, as any adaptive system the user model constructed should be explicit and accesible by the user and tutor. We will also see how we dynamically manage the user model in an adaptive learning environment by means of a ensemble of classifiers.

In this paper we present and discuss the results of the evaluation of an Intelligent Computer Assisted Language Learning (ICALL) system that operates over the Web. In particular, we aimed at evaluating the system along three dimensions: a) the effect of the intelligent features of the system on the learning outcome of students, b) the system's ability to provide individualized support to students that leads to more effective use of the system and c) the general usability and friendliness of the ICALL. To achieve this, we conducted an empirical study, where we compared the intelligent system with a non-intelligent version of it. The results of the study revealed that the students of the Web-based ICALL had gained more knowledge of the domain and had been able to interact with the system more effectively as compared to the students that had used the non-intelligent version of the system. However, the students of the intelligent version of the system found it more difficult and they needed more time to get acquainted with the system in comparison to the students of the non-intelligent system.