Analyzing the Total Household data of different states in India based on Census and evaluating their relative performance using Machine Learning

Abstract

As per the Census 2011 on Household data, Data will be collected based on Linguistics,
Economical, Social, Literacy rate, Nativity and so on. Each data represents a particular spatial feature. The
data considered in this problem refers to Households which are in good, livable and dilapidated conditions
and some more data collected also refers to the number of dwelling rooms available to the households on
tenure status.
In this work, a list of features are considered and the objective is to select an ideal number of features by
using a Predictive model. Thus, the model consists of features to be processed using Genetic Algorithm.
The idea here is to have put together a data frame containing the Good number of households and Livable
households of the population on whom to generate the best possible features for this problem. The specific
challenge is to select optimal number of states, at a total number of Good households with the predefined
sum which achieves better performance.