透過您的圖書館登入
IP:3.144.97.189
  • 學位論文

運用資料探勘於火警案件發生預測之研究

Research on the Predictions of Fire Incidents with Data Mining

指導教授 : 方鄒昭聰
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


近年來火警事件頻繁發生,由火警轉變為侵害人民財產、生命安全的火災過程中,事前掌握災害來源的相關性知識顯得相當重要。而在火警案件轉變為火災的過程中,若能有效率做到的資源整備與調度,使死亡與受傷人數降至最低,為本研究主要探討的內容。而本研究目的為分析現有火警通報資料庫的數據以建立「火警案件發生數量預測模型」,並評估所建立的預測模型,使模型的結果可以作為消防局相關人員的資源整備與調度之參考依據。 本研究採用文獻分析法找出過去文獻中不足的地方並加以改進。再者,透過深度訪談法,選取與處理火警案件之相關利害關係人進行訪談,改善其資料處理的過程與方法。最後,本研究證實時間序列分析運用於火警案件發生數量預測上具有良好適用性,並於研究中發現以一週作為預測火警發生頻率之期間是適當的,且以各個大隊為預測區域的適用性良好。

並列摘要


In recent years, fires occur frequently. When fires turn into threats to people’s property and safety, then getting to know the knowledge related to the source of damage in advance seems to be relatively important. This thesis is focus on when a fire turns into an actual damage, we could minimize the casualties by managing resources efficiently. This goal of thesis is to analyze database of fire cases and to build “prediction module of fire incident events” and to evaluate the module. The results of the module can be related references for fire fighters to manage and prepare resources. This thesis uses literature reviews to find the insufficient parts of previous documents and improve it. Furthermore, through expert interviews, selecting people who are related to the fire incidents to make interviews, by doing so, we can improve the processes and methods of data processing. Last, this thesis proves that time sequences which are applied to the predictions of fire incidents occurrences bring good effects. It discovers that using one week per a period to predict the frequencies of fire incidents in this research is suitable. Also, using each fire brigades as a prediction area is appropriate.

並列關鍵字

Data Mining Fire Incidents Time Series

參考文獻


翁士勛(2006)。建築物火災危險度預測模式之建構-以台南市為例。國立成功大學都市計劃學系碩博士班碩士論文。
林韋伶(2012)。建築物火警初期派遣之關聯性研究-以新北市為例。國立臺北大學資訊管理研究所碩士論文。
林君亭(2012)。城市火災安全評估之研究。國立臺北科技大學土木與防災研究所碩士論文。
Fayyad, U., Piatetsky-Shapiro, G., & Smyth, P. (1996). The KDD process for extracting useful knowledge from volumes of data. Communications of the ACM, 39(11), 27-34.
Han, J., & Kamber, M. (2006). Data Mining: Concepts and Techniques. Morgan kaufmann.

延伸閱讀