Application of Crowdsourcing to Generate Datasets of Public Procurement Processes

Abstract

In order to fight against Corruption, there are currently laws of free access to public information that favor transparency in many countries. In this opening scenario, information and communication technologies play an important role in facilitating the delivery of information that is made available to citizens seeking to enable greater control over the actions of government and civil servants. Nevertheless, in many cases, the information provided is aimed at human reading and interpretation and it is not published in machine-processable formats, a condition that hinders automatic processing using state-of-the-art computational techniques. This article presents a new method of data collection based on Crowdsourcing that allows generating data sets of public procurement processes from the unstructured information published on the Web. The results obtained show that our method allows the generation of reliable public procurement data sets in a scalable way whose scope is limited by the availability of data and the transparency of the country.