User Access Pattern Classification Scheme for Optimizing Web Directories

P. Rubini, K.Sudhakar, A.Anbarasan

Abstract


Web is the massive information source in the world. Information retrieval is the complex task in the web environment. Search engines handle the information retrieval process in two ways. They are query-based information retrieval and directory based information retrieval. Google provides the query based information retrieval model. All the information is fetched with respect to the user query value. Yahoo provides directory and query based information schemes. All information is arranged in the hierarchical domain order. Web directory is a hierarchical tree structure of domain and sub domain information. The web directories are classified into two types. Artificial web directories are constructed with reference to the web document contents. Real web directories are constructed with the usage data. Personalization can be applied on the real web directory environment. Objective Community Directory Miner (OCDM), Objective Probabilistic Directory Miner (OPDM) and Objective Clustering and Probabilistic Directory Miner (OCPDM) methods are applied in the existing web directory personalization schemes. The proposed system is designed to perform web directory optimization using the classification techniques. The probabilistic latent semantic analysis algorithm is used for the classification process. The fuzzy logic technique is used to enhance the PLSA scheme for weight optimization. The web directory optimization model uses ISP based user access logs.

 

 


Keywords: Personalization, Probabilistic Latent Semantic Analysis Algorithm, Fuzzy logic, Web directory optimization.


Full Text:

PDF


DOI: https://doi.org/10.26483/ijarcs.v2i5.730

Refbacks

  • There are currently no refbacks.




Copyright (c) 2016 International Journal of Advanced Research in Computer Science