Improved algorithmic techniques for mining and privacy preservation of Frequent itemsets

Abstract

It is but natural to have unprecedented advancements and evolution newlinein the field of database technology as it is the need of the hour to have a newlinemassive amount of data stored With a huge amount of data stored in files newlinedatabases and other repositories it is essential to develop powerful means for newlineanalysis and interpretation of such data and for the extraction of interesting newlineknowledge that could help in strategic decision making Data mining newlinetechniques have been extensively used for extracting non trivial information newlinefrom such massive amounts of data It is a powerful technology with great newlinepotential to analyze vital information and an essential step in Knowledge newlineDiscovery in Databases KDD which is the process of identifying valid newlinepotentially useful and ultimately understandable patterns in data It can make newlinea great contribution in many applications such as financial forecast medical newlinediagnosis consumer research marketing e commerce and so on and so forth newlineThis thesis proposes improved methods for mining and privacy preservation newlineof frequent itemsets newline newline