Improving Government Response to Citizen Requests Online

The Mexican constitution guarantees its citizens the right to submit individual requests to the government. Public officials are obligated to read and respond to citizen requests in a timely manner. Each request goes through three processing steps during which human employees read, analyze, and route requests to the appropriate federal agency depending on its content. The Mexican government recently created a centralized online submission system. In the year following the release of the online system, the number of submitted requests doubled. With limited resources to manually process each request, the Sistema Atencion Ciudadana (SAC) office in charge of handling requests has struggled to keep up with the increasing volume, resulting in longer processing time. Our goal is to build a machine learning system to process requests in order to allow the government to respond to citizen requests more efficiently. We develop models to help at each stage: to accept or reject incoming requests, distinguish between technical support questions to a help desk and substantive requests to the SAC, and send substantive requests to the appropriate federal agency. Given the office's needs, our system can automatically accept or reject 85% of all requests, automatically send 39% of requests to a technical helpdesk or to the SAC, and automatically routes 49% of all requests to the most appropriate federal agency. These models each operate at a different step in the manual request sorting process. Incorporation of our models into the request system would save the office of the president 3.8 man hours per day accepting and rejecting requests, 3.1 man hours per day deciding whether requests should go to the helpdesk, and 4.8 man hours per day routing requests to federal agencies. Together, this would allow the government to process 24.4% more petitions, allowing them to better address citizen needs with limited resources.