N 3 : NN Navigation Support System—Knowledge-Navigation in Hyperspace: The Sub-Symbolic Approach

A navigation support system based on the sub-symbolic approach to decide the appropriate navigation strategy is proposed, which can identify the user's needs and give some appropriate advice to the user in his/her exploring learning process and the training procedure of the neural network (NN) is described.

Abstract

A navigation support system based on the sub-symbolic approach to decide the appropriate navigation strategy is proposed. In exploring hyperspace, users often tend to be in undesirable states (e.g., get lost and so on). To improve these undesirable states, a sophisticated hypermedia system, which can identify the user's needs and give some appropriate advice to the user in his/her exploring learning process, was constructed. A hypermedia system with user-adaptive function based on an artificial neural network (NN) has been developed. A supervised NN was used as a navigation strategy decision module in that system. As a result of the evaluation of this system, the validation of the knowledge, which is learned by the NN, and the effectiveness of the navigation strategy, which is decided by the adaptive system, are shown. Hypermedia is a tool for user-driven access to information (Schneiderman, 1989). It provides an effective learning environment, where users can acquire knowledge by exploring the hyperspace in his/her own way. But, users often tend to get lost in the hyperspace. Often there appears the phenomenon of redundant information access. To improve these undesirable effects on users, many researches have been working to construct a sophisticated hypermedia system which can identify the user's interests, preferences, and needs and give some appropriate advice to the user in his/her exploring learning process. In this article, a user model that represents exploring activities in hyperspace (HS) and a mechanism based on a sub-symbolic approach, to decide appropriate navigation strategies is proposed. This article also shows the training procedure of the neural network (NN) and its result. Adaptive hypermedia is a flexible system which infers the learning goal or the current learning state, by using the exploring history, the structural characters of the hypermedia, and so on. As a result of its inference, this system changes its own performance to adapt to the user. Brusilovsky classified the adaptive hypermedia systems from the point of view of the methods and techniques of adaptation (Brusilovsky, 1996). According to his paper, the methods for adaptation are of two types. One is the content-level adaptation. This method changes the contents of the node, which the user will refer to in the next step. This type is also named the adaptive presentation system. The other one is the link-level adaptation. This method changes the links in the current node. This is also named adaptive navigation support system. The ELM-ART (Brusilovsky, 1996), which is an intelligent tutoring system, is one of the most famous adaptive presentation systems. Its adaptation was based on a specific domain knowledge wit h a symbolic approach. So, the generality of the adaptation mechanism and the flexibility of the functional extension of this system are not high. The Adaptive HyperMan (Mathe & Chen, 1996) is one of the most famous adaptive hypermedia systems using a method of the content-level adaptation. This system is an example of interactive adaptive hypermedia. This system bases its adaptability on the use of an Adaptive Relevance Network, which is made of a collection of personal data. This mechanism can store the users' interests. The exploring characteristics of each user are acquired through conversation with him/her. This system is useful for a situation in which a frequent interaction between the system and the user is needed. An adaptive educational hypermedia system based on the sub-symbolic approach has been developed. In cases of educational use, as the exploring aim of students is usually fixed by a teacher in advance, frequent interactions with the system may be obstacles for the free exploring learning of a student. This article also describes the training procedure of the NN and its result. The article is structured as follows. In the next section the indicators that evaluate the users' exploring activity in hyperspace and the goal and content of each navigation strategy is described. …