Abstract

The health and nutrition status of marginalized women with constraint economic resource in rural setting remains\na great challenge. In addition, poor nutritional knowledge, attitudes and dietary practices significantly contribute to the rapid\ngrowth of malnutrition among the rural women. In an attempt to re-architect the present scenario, it is necessary to employ\nthe interventions of science and technology in rural women community. To adapt the interventions of science and technology\npractically it is very much required to collect, analyze and interpret the data on health profile, family history, dietary habits\nand blood profile of an individual. Naturally this data is huge in volume, potentially high dimensional, absolutely diversified\nand structurally complex. To extract and study the insights of such data and segment the rural women efficiently it is\nnecessary to employ a capable data mining technique.\nTowards this goal, the investigators in the present paper present an Intelligent Data Classification System (IDCS). The IDCS\nis designed with four stages on the basis of classification technique with a comprehensive approach. Initially, the first stage\nof IDCS focusses on collecting the data from various resources on nutritional knowledge, attitudes and dietary practices of\nthe rural women systematically. Later it rightly preprocesses the data with the standard practices and prepare in suitable\nstructure to apply classification technique. Subsequently an intelligent classifier is modeled using learning algorithm which\ncan classify the rural women in desired number of groups with respect to nutritional levels. Finally, this system identifies\nsuitable interventions of science and technology like creating awareness about nutritional knowledge, arranging training\nworkshops to prepare the food without losing nutrients and ready to cook nutritious complementary food, promoting\nhomestead gardens and backyard poultry etc. appropriately to the respective rural women groups. This system helps to\nidentify the under nutritional women accurately and efficiently. This model reduces the error rate and shows the significant\nperformance. To validate the proposed system several experiments are conducted and results proven this are claimed in this\npaper.