Learning Contextual Behavior of Text Data
Coşkun Bayrak, Hemant Joshi · 2006
Understanding contextual behavior is very important in order to develop a context-aware retrieval system. This paper discusses the philosophy behind the development of the "evolutionary behavior of textual semantics" (EBOTS) system. The EBOTS system is retrieval oriented knowledge representation and management system. This paper proposes a formal model of correlation that can be combined with traditional local and global weighing schemes. Intuitive contextual behavior is studied as a part of proposed research work. Context retrieval based on semantic knowledge allows abstract queries to be defined, instead of exact word-based queries. The results of the context retrieval for a classic3 and time dataset using the EBOTS system have been discussed in this paper. The paper makes a contribution to the semantic knowledge representation and retrieval algorithms.