A Generic Pattern Matching Approach for Multiple Events

Abstract: Data mining is a highly evolving and developing frontier in data and information systems and their applications. Its a concept of extracting data from large data set and converts into understandable format. Automated prospective analysis and knowledge driven decisions make data more proactive. Pattern matching used for control flow of patterns, checks similarity between already stored templates with input template. Patterns are procedure for converting input data. The existing system pay-as-you-go style matching technique has poor discriminative feature were this technique depend mainly on time. So it will affect 1: n matching for events and when matching events with heterogeneous data it may result in duplication of data. In proposed system, as Generic Pattern Matching technique implies to validate similarity between two events. Number of events internally stores and maintains all the characteristics of events. Patterns are compared with all stored patterns with in the data set. Pattern matching enhances several events to improve the performance and efficiency of events. This is independent of time so 1 : n issue can be sorted out. It will also retrieve exact result without any approximate value.

Keywords: Pattern Matching, Generic Pattern Matching, Patterns, Automated prospective analysis.