G-Tweet: tool allowing to detect incidents on Twitter

Abstract

In this Paper we show how we have tackled the problem of bad opinions about a waste company services on Twitter [1]. We classified the tweets in two types, inci-dents: tweets that the company can fix, and opinions [2].

In the first phase our research consisted in development of computer algorithms capa-ble of detecting incidents published by citizens on Twitter, and in the second phase we managed the collected information to improve the service and control of citizen perception.