International Journal of Science and Research (IJSR) ISSN (Online): 2319-7064 Index Copernicus Value (2013): 6.14 | Impact Factor (2013): 4.438 Volume 4 Issue 9, September 2015 www.ijsr.net Licensed Under Creative Commons Attribution CC BY Mining GPS Data for Traffic Congestion Detection and Prediction Suhas Prakash Kaklij Department of Information Technology, Siddhant College of Engineering, Sudumbare, Pune University, Pune 412109, India Abstract: GPS data is available in the large amount, also for the devices having GPS a large amount data is being collected over time. The mining of this huge data is endorsed in discovery of the areas which face regular traffic congestion. User will have prior awareness of such locations which guide in deciding whether or not to go for that route. Avoidance of such routes will also assist in reduction of congestion of such locations. Also detected that the work which has been carried out till now in this field do not provide very precise and relevant results. The reason behind this is the no proper algorithm are selected and distinguished between on road and off road traffic. To deal with all this we proposed this system. This system will be structured and applied over GPS data i.e. data coming from devices like mobile phones, tablets, on board units etc. In the technique used in this system, these GPS data will be first cauterized using the K- means clustering algorithm. The clusters obtained are filtered out. On further processing these clusters a mining method of Naive bayes algorithm is used for mining for traffic Congestion detection and prediction Keywords: Traffic Congestion Detection, Traffic Jam Prediction, Traffic Tracking and Tracing. 1. Introduction Road network is biggest network widely used for Transportation. Each city has its Road network. Roads are used for daily transport not only for the people but for goods and many other things. The biggest problem now a days people facing is Traffic Congestion. The most of the congestion occur early morning or late afternoon because students and employees are going to their works and colleges so they also be late at traffic spot. People are not able to reach their work due to this traffic problem. As per the observation traffic congestion is dynamic in nature, it is not static. Means traffic congestion is variable as time passes and resources provided by current infrastructure are limited. In current emerging IT world we have lot of traffic data available with us in the different formats. With the use of this data we can get the flow of traffic information with respect to the location and time. This traffic information is important not only for current status of traffic but it can helps to analyze and predict upcoming traffic patterns. We can collect such information by processing GPS data. With use of 2G and 3G enabled GPS devices, huge set of data is collected with an average error of 2-15m [2]. These errors can further be decreased using some of the correction strategies such as map-based correction given in [2]. It is real time data which gives convenience to mine the traffic patterns of particular area. We can evaluate such data to get the traffic congestion patterns which in turn helps to identify the location where traffic congestion is possible. Prediction is also possible for traffic congestion of relative routes with respect to time. 2. Related Work Substantial amount of efforts experimented in the field of analyzing traffic patterns. H. Inose et al. In 1967, as given in [3], proposed how traffic signals are work systematically. It works for the minimization of delay in time of vehicles and providing appropriate and preferential offsets to the optimal graph in the road network. In 2002, Ashbrook et al., as given in [4], projected user consequential locations and end user activities using GPS data. As projected the city is divided in to clusters using K-means clustering further classified into a Markov Model. Thus, their work targeted on analyzing user GPS data to mine user momentous locations. As per the year 2010, Lipan et al. in [5], mined traffic patterns from GPS data concentrated from public transport. Their work focuses on observing bus schedules. Association guidelines are built on clusters where individual cluster has its own moderate speed. In 2011 [6] Mandal K and his team used probe vehicle technique for traffic congestion monitoring, Traffic information from probe vehicles has great potential for improving the estimation accuracy of traffic situations System tries to monitor the traffic flow pattern and then detect the congestion. As given in [7], Yao et al. proposed a speed pattern model which provide assumptions for traffic conditions and speed pattern with the help of machine learning. In 2013[1] Anand Gupta and his team proposed a groundwork for traffic congestion detection concentrating more on innovative algorithm which in turn cut down conflict of data for traffic Jam and Traffic signal. These efforts have given significant and helpful results. On the other hand, to the best of the authors’ discoveries, not much attention has been given to detection and prediction of traffic congestion with properly manipulating of on road & off road data, As well as the Conflict between the Traffic signal and Traffic Jam. Also no appropriate selection of mining and clustering algorithm. 3. Motivation Detecting traffic jam based on simple rules, such as using a probe vehicle technique for improving the estimation accuracy of traffic situations, velocity-based approach, and fuzzy logic might not handle the problem stated previously with great effect due to the following reasons Section headings come in several varieties: 1. Selection of no proper Clustering algorithms. 2. Use of less relevant mining methods. Paper ID: SUB158203 876