Sunida Srisuriyachai. Analysis of road traffic accidents in Nakhon Pathom province using data mining techniques . Master's Degree(Technology of Information System Management). Mahidol University. : Mahidol University, 2007.
Analysis of road traffic accidents in Nakhon Pathom province using data mining techniques
Abstract:
Data mining techniques were applied for the analysis of traffic accidents in
Nakhon Pathom. The data was obtained from Narenthorn Center, The Ministry of
Public Health, local police stations in Nakhon Pathom, and Traffic Report and
Accident Management System (TRAMS), Information and Communication
Technology Center, The Ministry of Transport.
A clustering algorithm (SimpleKMeans) and three classification algorithms
(J48, Naivebayes, and OneR) were the main focus. SimpleKMeans was used as the
main algorithm in the clustering experiment because it is simple and its results are
easy to interpret. Studies comparing classification algorithms have found Naivebayes
to be comparable in performance with J48 and OneR. Naivebayes have also exhibited
high accuracy and speed when applied to large databases. OneR generates extremely
fast a one - level decision tree expressed in the form of a set of rules that all test one
particular attribute. J48, C4.5 decision trees, decisions to classify data instances can be
extracted and used to build traffic accident profiles. By contrast, Naivebayes assign
classes to data instances without describing how the decisions have been made.
Considering the descriptive nature of the results and the classification performance,
it can be concluded that J48 is sufficiently useful and reliable.
The outcome of this research was traffic accident profiles, which could be
useful for investigating traffic accident problems in Nakhon Pathom.