Understanding the Dynamics of Crop Problems by Analyzing Farm Advisory Data in eSaguTM

Developing personalized information services for the effective delivery of functional information by exploiting latest advances in information and communication technologies is one of the active research area. At IIIT-H, Hyderabad, (India), efforts are being made to design a personalized agricultural advisory system called eSagu TM . In eSagu system, an agricultural expert give an expert advice to the farms based on the crop photographs and other related information. During 2004-05, the eSagu system was operated for 1051 cotton farms. Every farm received the expert advice once in a week. As a result, the data set of about 20,000 such advice texts has been generated. In this paper, we have carried out the text analysis experiments on the data set to understand the dynamics of farm problems. The graph of cluster size versus number of clusters on the advices of a particular day resulted into an exponential curve. In addition, it was observed that the group of farms which have received the same advice in particular week were not receiving the same advice in the subsequent weeks. The analysis shows that farming community needs a personalized advisory service in a regular manner as crop problems are dynamic and influenced by multiple factors. At the same time, there is a scope for improving the performance of eSagu by exploiting common problems of crops.