DATA MINING AND ITS APPLICATIONS FOR MODELLING RAINFALL EXTREMES

Abstract

Data mining is a new powerful technology which helps in extracting hidden predictive information (future trends and behaviours) from large databases and thus facilitating decision makers to make proactive, knowledge-driven decisions. In this paper, a brief overview of various data mining functionalities, and an extensive review of the works done on temporal data mining are discussed. Of the two frameworks of temporal data mining, one that of frequent episodes is discussed in detail by explicating the various algorithms developed so far. Also, a case study using one of the algorithms, Minimal Occurrences With Constraints And Time Lags (MOWCATL), for extracting the rules to explain the spatial and temporal variation for extreme events in India is discussed and the results are shown.