Karnaugh Map Approach for Mining Frequent Termset from Uncertain Textual Data

Abstract

In recent years, uncertain textual data has become ubiquitous because of the latest technology used for data collection. As the existing technology can only collect data in an imprecise way. Furthermore, various technologies such as privacy-preserving data mining create data which is inherently uncertain in nature. So this paper propose a frequent pattern mining technique for mining termsets from uncertain textual data. This technique has conducted a study on uncertain textual data using the Karnaugh Map concept. The paper describes the approach in a three step procedure. First, we review existing methods of finding frequent termsets from document data. Second, a new method UTDKM ( Uncertain Textual Data Mining using Karnaugh Map) is proposed for finding frequent termset from uncertain textual data. Finally, we carried out experiments to evaluate the performance of the proposed method. The experimental results demonstrate that the prominent feature of this method that is it requires only a single database scan for mining frequent patterns. It reduces the I/O time as well as CPU time.