Privacy Preserving Techniques in Data Stream and challenges
AbstractData mining gets valuable data from large amounts of knowledge. In latest, knowledge streams are new sort of knowledge, which are fully totally different from existing static knowledge. The characteristics of information streams are: data has timing preference; data distribution changes perpetually with time; the quantity of is large; knowledge flows in and out quickly; and immediate reply is important. Existing algorithmic program is intended for the static database. If the information changes, it'd be mandatory to rescan the complete dataset, that takes to a lot of computation time and providing late answer the user. The matter of privacy-preserving data processing has wide been studied and lots of techniques are realize. However, existing techniques for privacy-preserving data processing are designed for static knowledge bases and don't seem to be appropriate for dynamic data. Once got to perform computation at that point to providing privacy together so that the privacy preservation drawback of data streams mining is very huge issue. The success of privacy protective knowledge stream mining algorithms is measured in terms of its accuracy, performance, knowledge utility, level of uncertainty or resistance to data processing algorithms etc. but no privacy protective algorithm exists that outperforms all others on all attainable criteria. Rather, an algorithmic program could perform higher than another on one specific criterion. So, the aim of this paper is to present current situation of privacy protective knowledge stream mining framework and techniques.
How to Cite
KALIA, Aditi; GHASTE, Pallavi. Privacy Preserving Techniques in Data Stream and challenges. International Journal Of Emerging Technology and Computer Science, [S.l.], v. 3, n. 4, p. 11-15, oct. 2018. ISSN 2455-9954. Available at: <https://aspirepublishers.com/index.php/ijetcs/article/view/227>. Date accessed: 20 may 2019.