A CASE STUDY ON DETERMINING THE BIG DATA VERACITY: A METHOD TO COMPUTE THE RELEVANCE OF TWITTER DATA

1. This thesis project deals with the Veracity aspect of Big Data from the truthful vantage point. We are interested in estimating veracity from the data without relying on external information as it may not be feasible to collect and analyze external information.

2. There are some research conducted in this area such as Objectivity, Truthfulness and Credibility (OTC) model and measures of Veracity .

3. In this thesis, we propose entropy as another measure of Veracity and develop a method to associate entropy to topics. The entropy measure is compared against the previously defined measures.

4. There are two motivations for our approach. First, no measure available is 100% accurate in determining the truthfulness of data and so more measures are needed for validation. Second, entropy is used to measure the ambiguity in statements. We apply the measures to topics built on Twitter data.