Evaluating the Performance of Similarity Measures in Effective Web Information Retrieval

Information Retrieval (IR) manages recovering and showing data inside the WWW and online databases and furthermore looks through the web reports The quick development of site pages accessible on the Internet as of late, seeking applicable and coming data has turned into a pivotal issue. Data recovery is a standout amongst the most essential segments in web crawlers and their improvement would greatly affect enhancing the looking productivity because of dynamic nature of web it turns out to be much hard to discover applicable and late data. That is the reason an ever increasing number of individuals began to utilize centered crawler to get correct data in their uncommon fields today. The information retrieval field mainly deals with the grouping of similar documents to retrieve required information to the user from huge amount of data. The researchers proposed different types of similarity measures and models in information retrieval to determine the similarity between the texts and for document clustering. This research intends the study of genetic algorithm based information retrieval using similarity measures like cosine coefficient, jaccard coefficient, dice coefficient