Cluster Based Analysis and Consumption of Food Products in Targeted Public Distribution System
The Public Distribution System in India is 50 years old. At present it is being carried on as an anti-inflationary and antipoverty system. Tamil Nadu, the southernmost State in the country, is adopting the Universal Public Distribution System covering its entire population and supplying regularly rice, wheat, sugar, kerosene and other products like pulses, edible oil etc. The PDS is a centrally sponsored scheme that entitles beneficiaries to subsidized food grains every month. Several challenges have been identified in the implementation of PDS like (i) Targeting errors (ii) Large leakages or diversion (iii) The elimination of bogus cards and (iv) The problems in Fair Price Shops. This paper analyses and evaluates the problems and finds the possible solutions using the data mining techniques based on preprocessing and clustering. The K-means and K-harmonic means algorithms are combined to cluster the data based on the type of food commodities for rice and wheat.
