• SURJEET SINGH
• DEEPAK KUMAR

Traffic Management causes drivers' disappointment and costs billions of dollars yearly in lost time and fuel utilization. So as to beat such issues, this paper exhibits a component for Intelligent Transport Systems, which expects to identify and oversee activity clog. Restriction of movement on street systems is only lesser speeds, flourished outing time and flourished lining of the vehicles. At times when the quantity of vehicles increases than the capacity of the street, movement blockage occurs. In the urban areas of India, traffic clogging is an important issue of concern. Activity blockage happens when the request exceeds the approachable street limit. This way, the concentration is to diminish an opportunity to prepare, reroute and inform vehicles. Traffic flow forecast is the key purpose of Intelligent transportation frameworks investigate and in addition the vital condition for movement administration, control and direction. Presently conventional figure strategies and models incorporate nonparametric relapse demonstrate, exponential smoothing, time arrangement examination, counterfeit neural system, Kalman separating, movement re-enactment, Euclidian distance, dynamic activity task et cetera. The success of any framework depends on the forecast of traffic it has. The improvement in traffic can corrupt the execution of the framework. So it is important that precise estimation can be given.. So as to improve the expectation rate we propose a half breed method that uses the ARIMA model, KNN and Euclidean distance. Here and now activity stream is one of the center advances to perceive movement stream designs. In this exchange, in connection to the traits that the development of movement changes over and over, a fleeting activity stream forecast technique in light of a three-layered K-closest neighbor non-parametric relapse calculation is proposed. Especially, two screening layers in view of the similitude of shape were presented in K-closest neighbor non-parametric relapse technique, and the consequences of expectation were yield with the assistance of the weighted averaging on the complementary estimations of the likeness of shape separations and the strategy for most-comparable point remove modification.

Traffic Congestion Prediction (TCP), IoT, K- Means, FCM, KNN, Euclidean distance etc.

"A RESEARCH PAPER ON IMPROVING TRAFFIC CONGESTION PROBLEM BY IOT FOR SMART CITIES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 1, page no.692-699, January-2019, Available :http://www.jetir.org/papers/JETIR1901190.pdf

"A RESEARCH PAPER ON IMPROVING TRAFFIC CONGESTION PROBLEM BY IOT FOR SMART CITIES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 1, page no. pp692-699, January-2019, Available at : http://www.jetir.org/papers/JETIR1901190.pdf