Traffic Analysis by Using Random Forest Algorithm Considering Social Media Platforms
Kovuru Sridevi1, T. Ganesan2, B V S Samrat3, S Srihari4
1Kovuru Sridevi, Department of, Computer Science and Engineering, KLEF, Vaddeswaram, Guntur Dt (Andhra Pradesh), India.
2T. Ganesan, Department of Computer Science and Engineering, KLEF, Vaddeswaram, Guntur Dt (Andhra Pradesh), India.
3B V S Samrat, Department of Computer Science and Engineering, KLEF, Vaddeswaram, Guntur Dt (Andhra Pradesh), India.
4S Srihari, Department of Computer Science and Engineering, KLEF, Vaddeswaram, Guntur Dt (Andhra Pradesh), India.
Manuscript received on 24 March 2019 | Revised Manuscript received on 05 April 2019 | Manuscript Published on 18 April 2019 | PP: 620-625 | Volume-7 Issue-6S March 2019 | Retrieval Number: F03220376S19/2019©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Nowadays social media is an important platform for communication. People started utilizing social media platforms like Twitter, Facebook, WhatsApp, google+, Instagram, etc. People keep updating live incidents through these platforms. The live updates include weather details, traffic details at various places and current events in cities. In major metropolitan cities, traffic is a burning issue. People have to wait for quite a while in heavy traffic. Under such conditions, people started updating the traffic details and congestion details through these platforms. In this paper, social media data is used for analysis and prediction of traffic in the intervals of one hour. A web page has been created for making the estimation data available for users across the globe. Random forest algorithm is used for estimation of traffic based on the traffic congestion level three hours before the estimation period, the day of the week, whether the day is a holiday or not. 88% accuracy is achieved using this model. This model also presents an alternative route by comparing predicted traffic by this model in all the possible routes and suggesting the best possible route with minimum traffic for end users convenience.
Keywords: Data Mining, Prediction, Random Forest Algorithm, Social Media, Traffic Analysis.
Scope of the Article: Social Networks