A Methodology for Retrieving Datasets from Open Government Data Portals Using Information Retrieval and Question and Answering Techniques

Abstract

Public administration is one of the largest producers and collectors of data in several domains. Bearing in mind that one of the main pillars of public transparency is the release of social value through government data, and that this right remains a challenge for citizens, in this work we implemented a methodology that uses techniques and Answers and Information Retrieval to retrieve data sets and respond to questions from citizens on a Brazilian open government data portal. Using metrics of information retrieval evaluation, such as accuracy and recall, we obtained satisfactory results in applying the proposed methodology.