Pooja Bendre , Veena Kulkarni
Wikipedia has become a well-known information base in the most recent years since it is a general reference book that has a lot of data and in this way, covers a lot of various subjects. In this bit of work, we examine how articles and classes of Wikipedia identify with one another and how these connections can help during the query fetching process. Summing up is a procedure of recognizing significant data from a book. The procedures utilized by specialists are distinguished and converted into a lot of heuristic guidelines where the calculation is created dependent on the heuristic principles. A revolutionary graph-based totally textual content summarization version for normal single and multi-file summarization. The technique includes 4 processing degrees: parsing sentences semantically the usage of Semantic position labeling, grouping semantic arguments even as matching semantic roles to Wikipedia concepts, building a weighted semantic graph for every record and linking its sentences (nodes) through the semantic relatedness of the Wikipedia ideas. An iterative rating algorithm is then carried out to the document graphs to extract the most critical sentences information. Also proposed two methodology in which first approach was using two algorithms together graph based and DBSCAN which provide better result in few second interval. And second approach was to get the separate result of both of above algorithm. In this paper author purposed system has better result of summarization with comparative analysis and DBCSAN clustering technique proved to have efficient throughput and maintain semantic information of Wikipedia data.
Wikipedia; Graph-Based; DBSCAN; Parsing; Semantic; Summarize.