臺灣博碩士論文加值系統

嚴重之交通事故除了造成民眾生命財產之嚴重損失外，在國家整體生產力與社會形象中亦扮演著重要的角色；腳踏車運具介於休閒活動與交通工具之間，而未引起廣泛注意與研究；以民國96年為例，全國腳踏車事故件數為8536件，造成141人死亡及7982人受傷，死傷程度相當嚴重；加上國際油價飆漲，政府提倡節能減碳，腳踏車的使用受到重視，其交通事故數亦隨之提高。本研究針對腳踏車事故特性，運用參數型模式-羅吉斯迴歸模式(Logistic Regression Model)與非參數型模式-關聯法則(Association Rules)、多層次關聯法則(multilevel association rules)三個模式做分析，探討主要造成事故傷亡嚴重程度之主要因素與影響程度；然而，運輸安全相關之研究上，交通事故的發生被視為隨機事件，因此理論方面大多以統計方法分析，特別是參數型迴歸模式；而大多數的統計迴歸模式皆有適用的資料類型，且模式推導及建立之過程需有許多不同的假設條件，若違反此假設條件，則會造成錯誤之分析結果。關聯法則是資料探勘領域中發掘資訊之間關聯性之重要研究，在模式建立過程中無需任何假設，可在龐大資料庫中找出項目或屬性具關聯性之組合。因此，本研究使用民國94年至民國96年台灣三大都會區之腳踏車事故資料，研究之影響變數依當事者事故相關、道路幾何相關、環境相關與事故本身相關分成四類進行分析；研究結果顯示，腳踏車騎士特性、環境因素、道路型態及碰撞車之行動狀態為主要影響事故之因素。

Traffic accidents often result in serious traffic congestion which not only causes excessive delay for road users but a great loss of productivity for the society. Therefore, traffic accidents have been a great concern for the public. Among these traffic accidents, bicycle accidents have not yet drawn considerably attention from the general public because bicycles are not regarded as a regular transportation mode. In 2007, 8,536 bicycle accidents occurred in Taiwan resulting in 141 people killed and 7982 people injured. With the increase of gasoline prices, the popularity of bicycle uses is expected to grow rapidly in a short period. Therefore, there is an increasing need to have a better understanding the characteristics of bicycle accidents and the factors resulting in severe injury for the bicyclists. To explore the factors resulting in severe injury, regression analysis has been extensively applied. However, most regression models have their own model assumptions and pre-defined underlying relationships between dependent and independent variables. If these assumptions are violated, the model could lead to erroneous estimations of injury likelihood. The association rules, one of the most widely applied data mining techniques, has been commonly employed in business administration, industry, and engineering. Association rules do not require any pre-defined underlying relationship between dependent variable and independent variables and has been shown to be an effective tool, particularly for dealing with prediction problems. This study collects the 2005-2007 bicycle accident data of three major cities in Taiwan. The findings by the association rules indicate that the bicyclist characteristics, environmental factors, road characteristics, driver/vehicle action prior to accident are associated with injury severity of bicycle accidents.

致謝................................ I
摘要................................ II
Abstract............................III
目錄............................... IV
表目錄............................. VI
圖目錄..............................VIII
第一章 緒論.......................... 1
1.1 研究背景與動機.................... 1
1.2 研究目的......................... 4
1.3 研究範圍與對象.................... 4
1.4 研究架構與流程.................... 5
第二章 文獻回顧....................... 7
2.1 腳踏車事故相關文獻................ 7
2.2 交通事故相關文獻.................. 12
2.3 多元羅吉特迴歸模式相關文獻......... 15
2.3.1 多元羅吉特迴歸與交通事故......... 15
2.3.2 多元羅吉特迴歸與其他非交通事故..... 19
2.4 關聯法則的相關文獻................. 20
2.5 小結............................. 23
第三章 研究方法........................25
3.1 羅吉斯迴歸模式.................... 25
3.2 關聯法則......................... 26
3.2.1 關聯法則之模式.................. 26
3.2.2 關聯法則之步驟.................. 27
3.2.3 關聯分析....................... 29
3.2.4 關聯法則之例子說明.............. 30
3.3 多層次關聯法則................... 33
3.3.1 多層次關聯法則之例子說明........ 33
第四章 資料收集.......................37
4.1 樣本定義........................ 37
4.2 變數定義與說明................... 37
第五章 模式分析.......................41
5.1 敘述性統計分析................... 41
5.2 羅吉斯迴歸模式................... 50
5.2.1 模式結果與分析................. 50
5.3 關聯法則........................ 57
5.3.1 模式結果與分析................. 57
5.3.2 小結.......................... 61
5.4 多層次關聯法則................... 62
5.4.1 模式結果與分析................. 62
5.5 羅吉斯迴歸模式與多層次關聯法則之比較.. 66
5.5.1 羅吉斯迴歸模式之預測正確率......... 66
5.5.2 多層次關聯法則之預測正確率......... 67
5.5.3 小結......................... 68
第六章 結論與建議.....................69
6.1 結論............................ 69
6.2 建議............................ 71
參考文獻.............................72
附錄一(最小支持度為30%、最小可信度為90%之關聯法則結果)...77

一. 英文參考文獻

1. Al-Ghamdi, A. S. (2002), Using logistic regression to estimate the influence of accident factors on accident severity. Accident Analysis and Prevention 34, 729-741.
2. Al-Ghamdi, A. S. (2003), Analysis of traffic accidents at urban intersections in Riyadh. Accident Analysis and Prevention 35,717-724.
3. Andersson, A. L. and Bunketorp, O. (2002), Cycling and Alcohol. Injury, Int.J. Care Injured 33, 467-471
4. Agrawal, R., lmielinski, T. and Swami, A. (1993), Mining association rules between sets of items in large database. In Proceedings of the 1993 ACM SIGMOD International Conference on Management of Data, Washington DC, 207-216
5. Brian, H. Y., Lee, J. L. and Schofer, F. S. (2005), Bicycle safety helmet legislation and bicycle-related non-fatal injuries in California. Accident Analysis and Prevention 37, 93-102
6. Cheng, C. W., Lin, C. C. and Leu, S. S. (2010), Use of association rules to explore cause-effect relationships in occupational accidents in the Taiwan construction industry. Safety Science, 48436-444.
7. Choo,S. and Mokhtarian, P. L. , (2004), What type of vehicle do people drive? The role of attitude and lifestyle in influencing vehicle type choice. Transportation Research Park A 38, 201-222.
8. Chen, M. C. (2007), Ranking discovered rules from data mining with multiple criteria by data envelopment analysis. Expert Systems with Application 33,1110-1116.
9. Chae, Y. M., Ho, S. H., Cho, K. W., Lee, D. H. and Ji, S. H. (2001) , Data mining approach to policy analysis in a health insurance domain. International Journal of Informatics 62, 103-111.
10. Ekman, R., Welander, G., Svanstrom, L., Schelp, L. and Santesson, P. (2001), Bicycle-related injuries among the elderly-a new epidemic? Public Health 115, 38-43
11. Gkritza, k., Niemeier, D. and Mannering, F., (2006), Airport security screening and changing passenger satisfaction:An exploratory assessment. Journal of Air Transport Management 12, 213-219.
12. Hiselius, L. W. (2004), Estimating the relationship between accident frequency and homogeneous and inhomogeneous traffic flows. Accident Analysis and Prevention 36, 985-992.
13. Kim, J. K., Kim, S., Gudmundur, F. and Luis, A. P. (2007), Bicycleist injury severities in bicycle-motor vehicle accident. Accident Analysis and Prevention 39, 238-251.
14. Kelvin, K. M. and Yau, L. (2004), Risk factors affecting the severity of single vehicle traffic accidents in Hong Kong. Accident Analysis and Prevention 36, 333-340.
15. Karkhaneh, M., Naghavi, M., Brian H., Brent E., Jafari, N. and Saunders, D. (2008), Epidemiology of bicycle injuries in 13 health divisions Islamic Republic of Iran 2003. Accident Analysis and Prevention 40, 192–199.
16. Liu, B. S. (2007), Association of intersection approach speed with driver characteristics, vehicle type and traffic conditions comparing urban and suburban areas. Accident Analysis and Prevention 39, 216-223.
17. Liao, C. W. and Perng, Y. H. (2008), Data mining for occupational injuries in the Taiwan construction industry. Safety Science 46,1091-1102.
18. Lajunen, T. and Rasanen, M. (2001), Why teenagers owning a bicycle helmet do not use their helmets. Journal of Safety Research 32, 323-332.
19. Mongkolnavin, J. and Tirapat, S. (2009), Marking the Close analysis in Thai Bond Market Surveillance using association rules. Expert Systems with Application 36, 8523-8527.
20. Meulenersa, L. B., Leea, A, H. and Haworthb, C. (2007), Road environment, crash type and hospitalisation of bicyclists and motorcyclists presented to emergency departments in Western Australia. Accident Analysis and Prevention 39, 1222–1225.
21. McDonald, N. C. (2008), Children’s mode choice for the school trip : the role of distance and school location in walking to school. Transportation 35,23-35.
22. Parkin, J., Wardman, M. and Matthew P. (2007), Models of perceived cycling risk and route acceptability, Accident Analysis and Prevention 39, 364-371.
23. Pai, C.-W., Saleh W., (2007), Modelling motorcyclist injury severity by various crash types at T-junctions in the UK. Safety Science 43-51
24. Rasanen, M. and Summala, H. (1998), Attention and expectation problems in bicycle-car collisions: an in-depth study. Accident Analysis and Prevention vol.30, No.5, 657-666.
25. Schif M. A., (2008), Risk factors for pelvic fractures in lateral impact motor vehicle crashes. Accident Analysis and Prevevtion 40,387-391.
26. Sherry, A., Everett, R. A., Shults, L. C., Barrios, J. J., Sacks, R. L. and Oeltmann, J. (2001), Trends and Subgroup Differences in Transportation-Related Injury Risk and Safety Behaviors Among High School Students, 1991-1997. Journal of adolescent health 28, 228-234.
27. Sohn, S. Y., and Kim, Y. (2008), Searching customer patterns of mobile service using clustering and quantitative association rule. Expert System with Applications 34, 1070-1077.
28. Takubo, N. and Mizuno, K. (2000), Accident analysis of sports utility vehicles: human factors from statistical analysis and case studies. JSAE Review 21, 103-108.
29. Wesson, D., Spence, L., Hu, X. and Parkin, P. (2000), Trends in Bicycling-Related Head Injuries in Children After Implementation of a Community-Based Bike Helmet Campaign. Journal of Pediatric Surgery, Vol 35 No 5, 688-689
30. Zentner, J., Franken, H. and Lobbecke, G. (1996), Head injuries from bicycle accidents. Clinical Neurology and Neurosurgery 98, 281-285.
31. Zaki, M.J., Parthasarathy, S., Li, W., and Ogihara, M., (1996), Evaluation of Sampling for Data Mining of Association Rules. Technical Report 617, Computer Science Department, University of Rochester, NY, USA.



二. 中文參考文獻

1. 牛田一雄、高井勉、木暮大輔、陳耀茂，「資料採礦利用Clementine使用手冊」，鼎茂圖書出版股份有限公司，民國95年2月。
2. 麥可‧婓瑞(Michael J.A.Berry)、戈登‧林諾夫(Gordon Linoff)，「資料採礦-顧客關係管理暨電子行銷之應用」，數博網資訊股份有限公司，2001年1月。
3. 陳順宇，「多變量分析」，華泰書局，1998年7月。
4. 陳順宇、鄭碧娥，「STATISTICA手冊(I)基本統計」，華泰書局，1999年1月。
5. 陳順宇、鄭碧娥，「STATISTICA手冊(II)工業統計」，華泰書局，1999年3月。