Geospatial data mining for Agriculture pest management - a framework

One of the problems in addressing pest management is inadequate knowledge about the factors influencing pest population dynamics. To understand pest dynamics, scientists and researchers collect pest surveillance data and related agricultural operations regarding crops, farming practices, weather parameters, etc. These databases contain details of pest incidence, climatic, soil, agricultural practices and serve as repositories of information. Correlations between some of these factors and pest incidence based on statistical models have been developed to some extent. However, a functionally viable model for pest forecast and pesticide use is still needed by farmers for efficient and effective pest management. This paper describes the frame work of the pest management system using data mining techniques; which concentrate on providing, historical data, current and recommended pest and pesticide information and to be simulated pest models up to farm level. In this work, an attempt has been made to show how geospatial data mining integrated with agriculture including pest scouting, pesticide and climatological parameters are useful for optimization of pesticide usage and better management. The outcomes will reveal interesting patterns of farmer practices along with pesticide usage dynamics both in spatial & non spatial way and can help to make out the reasons for pest and pesticide exploitation.