Analytical Classification and Evaluation of Various Approaches in Temporal Data Mining

Abstract

Modern data bases have vast information and their manual analysis for the purpose of knowledge discovery is almost impossible. Today the requirement of automatic extraction of useful knowledge among large-capacity data is completely realized. Consequently, the automatic analysis and data discovery tools are in progress rapidly. Data mining is a knowledge that analyzes extensive level of unstructured data and helps discovering the required connections for better understanding of fundamental concepts. On the other sides, temporal data mining is related to the analysis of sequential data streams with temporal dependence. The purpose of temporal data mining is detection of hidden patterns in either unexpected behaviours or other exact connections of data. Hitherto various algorithms have been presented for temporal data mining. The aim of present study is to introduce, collect and evaluate these algorithms to create a global view over temporal data mining analyses. According to significant importance of temporal data mining in diverse practical applications, our suggestive collection can be considerably beneficial in selecting the appropriate algorithm.