Real-time natural language processing for crowdsourced road traffic alerts

Out of many issues we face in transportation today, road traffic has become the most crucial issue that directly affects our lives and economy. Despite of many implemented and progressing solutions, this issue seems to be remaining in a significant level in many countries and regions. Instead of fully relying on solutions provided by the authorities, public has come up with different approaches to deal with this problem. In this study we are focusing on one such solution which effectively uses a popular social networking service, Twitter. But still this crowdsourced traffic alert service has a limitation due to its nature; the natural language representation. We are trying to cope with this limitation by introducing a real time natural language processing solution to generate machine readable road traffic alerts. We observe many potentials of transforming this raw data into a machine readable format. An architecture that can effectively capture, transform and present this data has been proposed in this study and it has been implemented in a prototype level to demonstrate the uses of such a model. We expect to see extended models that can handle similar issues in future by combining multiple fields of information technology.