Allstate is an american insurance company, which has organised a recruitment kaggle competition in October 2016. At that time Allstate was developing automatic methods of predicting the cost of insurance claims.
A tabular dataset was given,having around 188k rows where each row is representing an insurance claim Kaggle dataset. The data contains 116 categorical and 14 continuous features. Due to privacy concerns there is no description available for the features,so we do not know what a particular feature represents in the real world.
The goal is to predict the value for the ‘loss’ column,which is a continuous value using various machine learning techniques.