User Related Rare Sequential Topic Pattern Using Data Mining

Abstract

Individuals utilize Internet for various purposes e.g. long range informal communication, blogging and so on concerning their unique situation. This prompts dynamic change in creation and circulation of archive streams over the Internet. This would challenge the theme demonstrating and development of individual points. In this paper, we have proposed Sequential Topic Patterns (STPs) mining over the distributed client mindful report streams and figure the issue of mining User Aware Rare Sequential Topic Patterns(URSTPs) in archive streams on the Internet to discover uncommon clients. They are for the most part uncommon and rare over the Internet. For URSTPs mining we have to perform three stages: pre-preparing to separate points, creating STPs, deciding URSTPs by irregularity examination of STPs. The test can be performed on both genuine circumstances (Twitter) and manufactured informational collections. In the proposed work, we have concentrated on engineered datasets.