International Journal of Artificial Intelligence in Education: The European Journal on Artificial Intelligence

This research describes an Intelligent Collaborative Support System(ICSS) that supports a collaborative effort by analyzing and modifying the collaborative process dynamically while employing a web-based interface. Based upon principles rooted in Computer-Supported Cooperative Work (CSCW), Intelligent Tutoring Systems (ITS) and Cooperative Learning (CL), this system extends the Group Leader paradigm to assist students working together in collaborative groups. Discussion skills are supported by examination of sentence openers chosen from a menu, keywords found in free-text sentence closers, student and group models, and historical database files. Groups are categorized and guided toward the optimal category of a high-performing cooperative group with positive interdependence. The use of the dialogue designated as creative conflict is mediated by an agent to assist in formulating a constructive discussion, serving as an instructional tool. Conflicts in the categorization of discussion skills exhibited in the sentence openers versus the sentence closers are resolved. Implemented in Java, the Group Leader module interacts with intelligent agents as it guides students working in groups remotely in the cross-platform environment provided by the Java Virtual Machine.

Several hypermedia learning environments have been developed in the last years, with the objective of helping students to acquire specific concepts in a given domain, as well as problem-solving abilities. However, there are still many different and sometimes conflicting claims about learning effectiveness using such environments. We present an empirical evaluation of a learning hypermedia with a tutorial component that exploits Artificial Intelligence techniques. This hypermedia, called Logiocando, has been designed for use by a special category of user, namely children of the fourth level of primary school (9–10 years old), to help them to learn basic concepts of logic. This category of user demands special attention to usability of the hypermedia. For this reason, design and development of the hypermedia have been carried out following a learner-centred methodology, in order to build a system that satisfies clear usability objectives. The aim of the study herein reported was to evaluate the learning effectiveness of Logiocando and to estimate the difference between two approaches: computer-based using a hypermedia system, and traditional, namely a typical lesson in the classroom. The results have shown that the hypermedia can be considered a valid support in the process of learning and deepening a topic.

This article describes the ExpertCop tutorial system. ExpertCop is a geosimulator of criminal dynamics in urban environments that aims to train police officers in the activity of preventive policing allocation. ExpertCop intends to i) help students comprehend the task of resources allocation and identify the variables involved; ii) induce students to reflect about their actions related to resources allocation; and iii) help students understand the relation between preventive policing and crime. In ExpertCop, the students(police officers) configure and allocate an available police force according to a selected geographic region, and then follow the simulation process. The students interpret the results with the aid of an intelligent tutor, the Pedagogical Agent, by observing how crime behaves in the presence of the allocated preventive policing. The interaction between domain agents representing social entities such as criminals and police teams allocated by the students drives the simulation. Assisting the user, the Pedagogical Agent aims to define interaction strategies between the student and the geosimulator in order to make simulated phenomena better understood. The innovative character of the work is in the strategy adopted by the Pedagogical Agent in order to assist the user in understanding the simulated model. It seeks to explain the model at the macro-level (global or emergent behavior) and micro-level (behavior of the agents individually). The evaluation of the system in training held with students of the domain has shown that the system attains its intended pedagogical goals.