Anteneh, Fentahun (2011) Mining Road Traffic Accident Data for Predicting Accident Severity to Improve Public Health – Role of Driver and Road Factors in the Case of Addis Ababa. Masters thesis, Addis Ababa University.
![[img]](http://thesisbank.jhia.ac.ke/style/images/fileicons/application_pdf.png) |
PDF (Mining Road Traffic Accident Data for Predicting Accident Severity to Improve Public Health – Role of Driver and Road Factors in the Case of Addis Ababa)
Anteneh, Fentahun.pdf - Accepted Version Restricted to Repository staff only Download (1MB) | Request a copy |
Abstract
Road traffic accidents are among the top leading causes of deaths and injuries of various levels in Ethiopia. One of the solutions to reduce the problem of traffic accident is finding the causes through research, and data mining is one research tool in finding the causes of traffic accidents. The objective of this study is to identify and investigate drivers‟ and road factors that contribute to the cause of accident and to develop traffic accident prediction model. In this research an attempt is made to apply the decision tree and rule induction predictive data mining techniques in major driver and road factors for car accidents and identify hidden patterns in the accident data set. To achieve this goal: the CRISP-DM 1.0 standard data mining methodology is adopted and the WEKA data mining tool is used to implement the ID3 , J48 and PART algorithms. The data for this research is the RTA data of the years 2005-09 collected from the Addis Ababa Road Traffic Control and Investigation Department and local researchers. After preprocessing a total of 16,710 RTA records are used for building the models. Various experiments are made iteratively by making adjustment of the parameters and using different number of attributes to come up with a meaningful output. Major factors of drivers and roads are identified and rules are generated using J48 decision trees and rule induction (PART algorithm). The comparison of the models using WEKA's experimenter showed that J48 slightly outperforms ID3 and PART algorithms. In addition, the determinant factors of drivers and roads that cause road accidents are identified; these are LicenceGrade, subcity, RoadJunction, TypeofRoad, and LightCondition. In many data mining researches on traffic accidents, decision trees and neural networks are widely used but in this study rule induction and decision trees are used to built the different models that can solve the problem of public health in the society.
| Item Type: | Thesis (Masters) |
|---|---|
| Subjects: | H Social Sciences > HE Transportation and Communications R Medicine > RA Public aspects of medicine > RA0421 Public health. Hygiene. Preventive Medicine T Technology > T Technology (General) T Technology > TE Highway engineering. Roads and pavements T Technology > TN Mining engineering. Metallurgy |
| Depositing User: | Selom Ghislain |
| Date Deposited: | 16 Aug 2018 08:43 |
| Last Modified: | 16 Aug 2018 08:43 |
| URI: | http://thesisbank.jhia.ac.ke/id/eprint/4803 |
Actions (login required)
 |
View Item |