A Correlation-Based Semantic Model for Text Search

Abstract

With the exponential growth of texts on the Internet, text search is considered a crucial problem in many fields. Most of the traditional text search approaches are based on “bag of words” text representation based on frequency statics. However, these approaches ignore the semantic correlation of words in the text. So this may lead to inaccurate ranking of the search results. In this paper, we propose a new Wikipedia-based similar text search approach that the words in the texts and query text could be semantic correlated in Wikipedia. We propose a new text representation model and a new text similarity metric. Finally, the experiments on the real dataset demonstrate the high precision, recall and efficiency of our approach.

The work is partially supported by the National Natural Science Foundation of China (Nos. 61322208, 61272178, 61129002), the Doctoral Fund of Ministry of Education of China (No. 20110042110028), and the Fundamental Research Funds for the Central Universities (No. N120504001, N110804002).