Cognitive Location-Aware Information Retrieval by Agent-Based Semantic Matching

Abstract

This paper proposes semantic TFIDF, an agent-based system for retrieving location-aware information that makes use of semantic information in the data to develop smaller training sets, thereby improving the speed of retrieval while maintaining or even improving accuracy. This proposed method first assigns intelligent agents to gathering location-aware data, which they then classify, match, and organize to find a best match for a user query. This is done using semantic graphs in the WordNet English dictionary. Experiments will compare the proposed system with three other commonly used systems and show that it is significantly faster and more accurate.