A SURVEY ON TRUTH DISCOVERY METHODS FOR BIG DATA

Abstract

Increasingly large numbers of embedded Smart phones, Sensors, PCs, Tablets, Computers connected to network, internet Medical data, Business transactions, Data are captured by sensors, Social media/networks, Banking, Marketing, Government data, etc are generating enormous amounts of unstructured data. This data creates new opportunities to extract more value for the areas for which it is needed. Recently, the Big Data is a challenging one by a dramatic increase of data from the physical world. One important property of Big Data is its wide variety, i.e., data about the same object can be obtained from various sources. Most of the time sources provide conflicted data for the same object. It is the challenging one to identify the “True Information” from the noisy information. To overcome such difficulties, Truth discovery methods are developed by estimating weight of the each source that is reliability of the sources. This survey focuses on the methods which are used to find out the true information from the conflicted data and comparisons of methods are used to select the appropriate method based on the types of data. Key Words: Truth Discovery, Big data, Jaccard distance, Levenshtein distance