Information reconstruction via discrete optimization for agricultural census data

In the case of large-scale surveys, such as a Census, data may contain errors or missing values. An automatic error detection and correction procedure is therefore needed. We propose here an approach to this problem based on Discrete Optimization. The treatment of each data record is converted into a mixed integer linear programming model and solved by means of state-of-the-art branch and cut procedures. Results on real-world Agricultural Census data show the effectiveness of the proposed procedure.