Learning Paths Matching in a Hypertext Tutoring System

• Main Authors: Fu-Xiong Lin, 林富雄
• Other Authors: Yu-Tai Ching
• Format: Others
• Language: en_US
• Published: 1994
• Online Access: http://ndltd.ncl.edu.tw/handle/53289841374188898580

碩士 === 國立交通大學 === 資訊科學學系 === 82 === In a network oriented hypertext tutoring system, we record users'' browsing trails. They are called learning paths. Given a path, we want to find the most similar one from the others. Hypertext is usually represented by a directed graph (digraph) and a learning path is represented by the path generated by traversing the digraph. The similarity of two learning paths is defined by longest common subsequence (LCS). In an acyclic hypertext, we develop a method to compute the LCS efficiently. Many matching techniques and attributes of a learning path are introduced. Attributes of learning paths are used to define metrics. Once we know the distance of any pair of learning paths, we can get students'' learning patterns by applying clustering methods.