A Semantic and Feature Aggregated Information Retrieval Technique for Efficient Geospatial Text Document Retrieval
R. Uma and K. Muneeswaran
Processing the normal text is quite easier and the information can be efficiently retrieved. There are various algorithms have been already proposed for normal text retrieval. Whereas retrieving the geospatial information are very complex due to nature of the data. It needs supplementary processes to be performed. Since geospatial data contains complex information like location and direction. To effectively handle the geospatial queries, a novel Semantic and Feature Aggregated Information Retrieval is proposed in this paper. Preliminary, there are four steps need to be perform, they are Clustering, Indexing, Retrieval and Ranking. Also, Context based QueryWeighting (CQW) approach is proposed to cluster the documents present in the corpus and indexing is based on multilevel hashing. Feature Probability and Density (FPD) technique is utilized to retrieve the document which matches the user query information. The Semantic Density (SD) technique is used to rank the retrieved documents. The experimental results shows that the proposed SFAIR technique provides better results than the existing technique.
Keywords: Clustering, context based query weighting (CQW), feature probability and density (FPD, indexing, information retrieval, ranking and semantic density (SD)