DataMining

The objective is to predict poverty on a household level. We are given data on the individual level with each individual having unique features but also information about their household. In order to create a dataset for the task, we'll have to perform some aggregations of the individual data for each household. Moreover, we have to make a prediction for every individual in the test set, but "ONLY the heads of household are used in scoring" which means we want to predict poverty on a household basis.