Cooperative Agents in Web-Based Distance Learning

In the last few years, we have observed an explosive growth of multimedia computing, communications, and applications. This revolution is transforming the way people live, work, and interact with each other, and is impacting the way businesses, education, entertainment, and health care are operating. Due to the opportunities provided by the Internet, more and more people are taking advantage of distance learning courses. During the last few years, enormous research efforts have been dedicated to the development of distance learning systems (Katayama & Kambayashi, 1999; Nakabayashi et al., 1997). Consequently, many large projects such as the CALAT Project (www.calat.com), CALsurf (http://webbase.ntts.co.jp),, WebCAI (http:// iclap.ce.nihon-u.ac.jp/~webcai_), The University of The Air (http://www.u-air.ac.jp/, and WIDE University (http://www.sfc.wide.ad.jp/soi/contents.html) have been established (Ogawa, Ijiun, & Murai, 1999). Recently, several distance learning systems that consider the learner’s capability and understanding have been proposed (Matsumoto, Nakayasu, Morita, & Kamejima, 1999; Tamaki et al., 2000). In Matsumoto et al. (1999), an evaluation system of historical data based on learning environment and supported by educational software record is proposed. In this system, reappearance and analysis are carried out only for historical learning data and not for real-time ones. In Tamaki et al. (2000), a multimedia assisted education system with individual student advancement control (MESIA) is proposed. The system is able to keep the teacher operating cost low and to offer fine education through the cooperation of Computer-Assisted Instruction (CAI) and teacher. The system is able to recognize the learners who need assistance, but its main purpose is to support the teacher, not the learners. In order to offer a suitable and efficient study for learners, in this work, we propose a Web-based distance learning system using cooperative agents. The purpose of our system is to deliver appropriate study materials by judging the learner’s degree of understanding. The main elements of the proposed system are the agents, which play the teacher’s role and, based on the learning history, analyze the learner’s degree of understanding. To evaluate the proposed system, we developed several experiments and surveys. The evaluation shows that our system can achieve a good delivery of study materials for different learners. Also, by using new features, such as mental action of color and the competition with other learners, the system can further increase learning efficiency. The article is organized as follows. The system design is introduced in the next section. Then, we deal with data processing and present how our system stimulates the learner’s volition. In the following section, we discuss the experimental results. Finally, we give some conclusions and future work.