Positive improvements in query recommendation system using query flow graph with metaheuristic techniques

Abstract

An application in techniques of data mining related to web data is known as web mining The current work on such data mining is generally to get knowledge at the level of queries as well as the documents that are assessed in their response. An insight is gained through the logs as to how the search engine can be used or what the interests of the users are, the sequence of the user pursued accesses known as web access pattern is very useful and interesting in practice. This is normally used for the prediction of the navigational behavior of users Here the web log mining is used to better the performance of the search engines by making use of knowledge that is mined Information Retrieval (IR) which is the process by which data mining is done needs to extract patterns which should be valid and should bring out relationships that are obtained from huge sets of data Query recommendations are a prominent feature of modern search engines Query recommendations serve several purposes correcting possible spelling mistakes, guiding users through their information seeking tasks allowing them to locate information more easily and helping them explore other concepts related to what they are looking for Improving the provided service through a search engine is much useful by finding a similarity measure among queries To progress the user search knowledge via search engines has been frequently used by the users to interact information. Suggesting query information which is equal to user query is the query recommendation and it should be such that it improves recovery performance as a method An out come with an essential tool for query log mining is the Query Flow Graph QFG Datasets like ImageCLEF 2005 Internet Movie Database IMDB and four universities are selected for experiments newline