Risk Distribution and Validation of Data in Passport Data Analysis Using Cluster Analysis

Sucheta Gulia, Dr. Rajan Vohra

Abstract: This paper presents a comprehensive statistical experiment to identify the sensitive office from the passport database. Data mining is the process of finding the meaningful patterns, correlations among dozens of fields that lie hidden within very large databases. The presented work focus on implementing different data mining approaches on passport database which is the primary data taken from passport offices. This paper combines two major approaches to provide profiling via clustering and validation of data using classification. According to defined approach, clustering is implemented to profile offices according to their risk identified. The design of experiments software named Weka Tool is used for making clusters using attributes place of issue and risk type dataset by using simple k means algorithm. With the simulation and analysis results, identify the sensitive centre which contribute to high risk score

Keywords: Clustering simple k-means, Profiling, Distribution, Validation, Weka tool