Indian Journal of Science and Technology

Abstract

Background/Objective: Creating queries for a single search term and identifying the viable solutions for the query are the two specific problems in retrieving the data. To resolve this issue, an effective information fusion technology should be provided to obtain effective results. This paper presents a method for resolving conflicts and eliminating duplicates with increased accuracy. Methods/Statistical Analysis: Universal wrappers are designed to retrieve the actual information from the heterogeneous data sources. The process of getting input itself is modified such that the retrieved results are relevant to the context. Ranking and duplicate eliminations are done accordingly to refine the obtained results to the user. Findings: Experimental results show that the improved accuracies of the data being fetched and with reduced conflicts and duplicates. This work uses major data sources from Google, New York Times and other offline data sources. By applying the proposed data retrieval techniques, the produced data is consistent by the help of wrappers. The proposed approach improves the data consistency which is relatively better than the existing technique. Finally, this proposed research work concludes that it is used to identify and resolve the conflict data and delivers the consistent data to the users in a ranked manner. Applications/Improvements: To create a unified repository which can be used for knowledge mining and warehouse based analysis of existing data and retrieve the result.
Keywords: Conflict Identification, Conflict Resolution, Data Retrieval, Ranking, Wrappers