Semantic Retrieval of Web Documents using Topic Modeling Based Weighted Nearest Neighborhood Technique

Abstract

Information retrieval systems are used to retrieve documents based on the keyword search. Semantic-based information retrieval is beyond standard information retrieval and uses related information to get the documents from the corpus. But semantic retrieval based documents is not efficient enough in real time. Content from the user’s profile is used for searching the web documents. The documents which exactly matches the user requirement is retrieved and it improvises the personalized retrieval. In this paper, a methodology based on topic modelling is proposed to determine the retrieval of information for user to increase the accuracy of documents using Latent Dirichlet Allocation (LDA) and Weighted Nearest Neighbor (WNN) models. LDA model is developed to retrieve documents based on topics. The topic based retrieval is improvised using personalization technique which uses WNN model. Experimental analysis on building personalization and semantic retrieval of documents shows the improved precision compared to existing topic modeling.