Approach of Text Search Based on Semantic Parsing Model

Abstract

Keyword-based search is one of the most important technologies of text search. But with the development of World Wide Web, it is not enough only relying on matching the form of keywords. This paper introduced a semantic parsing model constructed above a symbolic system of concepts for understanding natural language. Based on the model a sentence can be mapped into the corresponding semantic category. An approach named 'hypothesis testing' was used to analyze a sentence according to the conceptual knowledge base and rules. All rules were formalized and executed in computer program. There is also a semantic model for analyzing text. Therefore, both sentence and text can be mapped into semantic structures. A text search prototype based on the semantic model represented more accurate result by comparing the conceptual structures between search condition and text.