A Survey on Securely Mining UARSTP with Recommendation System in Document Streams

Textual documents made and appropriated on the Internet are perpetually changing in different structures. The vast majority of existing works are committed to theme demonstrating and the development of individual topics, while consecutive relations of topics in progressive documents distributed by a particular user are overlooked. In this paper, so as to describe and recognize customized and abnormal behaviors of Internet users, we propose Sequential Topic Patterns (STPs) and plan the issue of mining User-mindful Rare Sequential Topic Patterns (URSTPs) in document streams on the Internet. They are uncommon overall yet generally visit for particular users, so can be connected in some real-world situations, for example, real-time observing on abnormal user behaviors. We exhibit a gathering of algorithms to explain this imaginative mining issue through three stages: preprocessing to separate probabilistic topics and distinguish sessions for various users, creating all the STP applicants with (expected) bolster values for every user by example development, and selecting URSTPs by making user-aware rarity analysis on determined STPs.