Geographical knowledge discovery applied to the social perception of pollution in the city of Mexico

Nowadays, experts and citizens at large are keen to express their opinions using social networks on many issues, this generating a new form of participatory democracy. The research presented in this paper proposes a preliminary research work that combines semantics processing and machine learning to derive geographic and semantic knowledge implicitly derived from the perceptions and opinions as expressed by social networks, digital media and institutional data. The results are mapped to the geographical structure of the city in order to study differences and commonalities at the neighborhood level. The whole approach is applied and illustrated in the context of the city of Mexico and pollution perception as a case study. The figures that emerge show evidence of a significant impact of the structure of the city over the way citizens perceive pollution.