Social networks analysis with use of machine learning to predict vehicles traffic in urban areas

• Grant number: 17/00406-8
• Support type: Scholarships in Brazil - Scientific Initiation
• Effective date (Start): April 01, 2017
• Effective date (End): December 31, 2017
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal researcher: Roberto Marcondes Cesar Junior
• Grantee: Lucas de Carvalho Dias
• Home Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

The area of urban informatics has developed intensively in the last years. Data science techniques are being created to analyse problems such as urban violence, intense traffic, traffic accidents and energetic consume. The present project fits in this context. Using data mining techniques information from social networks, traffic apps and news fonts will be extracted, filtrated and formatted. By applying machine learning methods in this data conclusions about how traffic is affected depending on the specifications of events that happen in the city (of human or natural nature). Still with machine learning techniques, these conclusions will be used to make predictions about how future events will interfere in the traffic flow. Additionally, the present project intends to use data from more than one social network and also from another fonts in the internet (like meteorological data). (AU)