Decision Tree Classification for Traffic Congestion Detection using Data Mining

Authors: , ,

Abstract: Traffic congestion could be a challenge in Gauteng province of Republic of South Africa and it’s a negative impact on the economy of this province therein services and merchandise aren’t being rendered on time. Holdup affects the standard of lives of Gauteng residents and guests alike. This information was used for constructing the traffic flow prediction models. To propose a way to spot road holdup levels from rate of mobile sensors with high accuracy .According to Human perceptions holdup levels divided into 3 levels: lightweight, heavy, and jam. The ratings and rate were fed into a choice tree learning model (J48).The slippery windows technique to capture the consecutive moving average velocities, that was referred to as a moving pattern .Then road users’ judgments and connected info were learned utilizing a decision tree model (J48) classification algorithmic program. The evaluations disclosed that the decision tree model. The achieved Associate in Nursing overall accuracy as high as 91% with a precision as high as 94%.

WSEAS Transactions on Computers, ISSN / E-ISSN: 1109-2750 / 2224-2872, Volume 17, 2018, Art. #27

PDF
XML
Cite
×
Citation Tools
WSEAS AMA
R. Sujatha, S. Srinithibharathi, S. Subhapradha, "Decision Tree Classification for Traffic Congestion Detection using Data Mining," WSEAS Transactions on Computers, vol. 17, pp. 220-226, 2018, DOI:
R. Sujatha, S. Srinithibharathi, S. Subhapradha. Decision Tree Classification for Traffic Congestion Detection using Data Mining. WSEAS Transactions on Computers. 2018;17:220-226.
Copy to Clipboard
Citation copied to Clipboard

Certification

---