透過您的圖書館登入
IP:18.118.45.162
  • 期刊

以資料探勘技術應用於學生水域安全之研究

Study on Application of Data Mining in Student Safety during Water Activities

摘要


臺灣處於亞熱帶島嶼區,四面環海,溪流、湖泊眾多,盛夏熱暑時,學生參與水域活動頻率漸增,因過於熱衷水域活動的樂趣,卻忽略水域潛藏的危險,以致溺水事故頻傳。本研究旨在探討學生之水域活動安全,研究工具採懷卡托智慧分析環境(WaikatoEnvironment for Knowledge Analysis, WEKA)軟體進行資料探勘,包含資料進行預處理、分類演算法、關聯規則等,分析結果得知學生最容易發生溺水的地點於近海處,從5月起至8月期間,特別集中在7月,學生為發生溺水事件的主要族群,溺水的主要原因以戲水居多,溺水者身分以國中生為最高。而溺斃事件男性同學為女性同學的3.5倍。因此,從放暑假前兩個月起,應加強宣導與強化學生水域安全知能,推動學生正確的水域運動觀念,以降低溺水事件發生機率。

並列摘要


Located in a region of tropical islands, Taiwan is surrounded by ocean and has numerous beaches, rivers and lakes. In summer, students more frequently participate in water activities and greatly enjoy the activities; however, they often neglect the danger of water activities. Hence, drowning happens all too frequently. This study treated students as subjects and probed into student safety during water activities. WEKA was used for data mining, which included data pre-processing, classification algorithms, and association rules. According to the analytical result, students are the main group susceptible to drowning, and students tend to drown in the near-sea area from May to August, particularly in July. The main cause of drowning is water activity. Drowned students are mostly junior high school students, and male victims outnumber female victims by 3.5 times. Thus, two months before summer vacation, it is important to reinforce students' education and knowledge of water safety and introduce the correct concept of water activities, in order to avoid drowning accidents.

延伸閱讀