行動電話詐欺管理之研究

CCUR > College of Business > Department of Information Management & Graduate Institute of Information Management > Thesis > Item 987654321/27532

Please use this identifier to cite or link to this item: https://irlib.pccu.edu.tw/handle/987654321/27532

Title:	行動電話詐欺管理之研究 A Study on Mobile Phone Fraud Management
Authors:	劉健安
Contributors:	資訊管理研究所碩士在職專班
Keywords:	行動電話詐欺詐欺管理決策樹貝氏機率類神經網路資料挖掘 Mobile Phone Fraud Fraud management Decision Tree Bayes Classifier Neural Network Data Mining
Date:	2006
Issue Date:	2014-06-20 10:42:15 (UTC+8)
Abstract:	本論文針對行動電話詐欺作出定義，研究詐欺者利用行動電話詐欺行為之原因，探討詐欺者之詐欺行為對行動電話業者所產生之影響，並且將國內外詐欺類型使用技術性及非技術性加以歸類。本論文提出利用人工智慧加以輔助分析偵測結果，並進而提出行動電話詐欺管理流程、管理策略之建議，期能幫助行動電話業者能夠妥善管理詐欺行為，降低電信詐欺所帶來的營業額損失及呆帳情形，讓欲以行動電話當成犯罪途徑的詐欺者有所警惕。本研究的分析結果顯示，以預測機率越高的預測方法將可疑名單進行排序，會得到更好的預測結果。本研究進一步發現若將兩種預測方式加以結合排序，會得到比使用單一種預測方法更佳的預測結果。故本研究證明以人工智慧輔助排序，再由客服人員針對可疑用戶名單進行訪談，的確可大幅提升電信詐欺管理的效能及效率。最後本論文提出人工智慧輔助之詐欺行為分析模式。若電信業者能找出更多不同的詐欺類型，篩選出相同類型的用戶，套用詐欺行為分析模式使用人工智慧輔助排序，再將排序過後的結果交由客服人員進行訪談確認，對於行動電話業者減少損失的程度將更為可觀。 The study made a definition on mobile phone fraud management. The study analyzed on the reasons why those fraudsters want to take advantage of mobile phone fraud and discussion the influence of these fraud actions for mobile phone operators. The study also classified the national and international fraud categories by the usage of technical and non-technical. The study addressed that use artificial intelligence to assist the results analysis. Fur-thermore, it addressed the process of fraud management, the suggestions for manage-ment strategy. We hope the study can help mobile phone operators to deal properly with fraud actions and reduce the profit lost and bad debts. It can also help to alert those fraudsters who want to crime by the way of mobile phone. The study analysis result showed that via sorting the suspect list with high probability forecast could get the accurate result. From the study results, we can also found that via sorting these two combined forecasts would get accurate forecast than single one. Therefore, the study can prove that sorting by artificial intelligence first, then we can ask customer service staffs to interview with those suspect users. With these two steps, mobile phone operators can improve the competence and efficiency in fraud manage-ment. In conclusion, the study addressed that fraud action modes analysis via artificial intelli-gence. If mobile phone operators could find out more different fraud categories, sift out same user categories, and then sorting these categories together with fraud action analy-sis modes and artificial intelligence. Pass those results for customer services staffs for further interview. Following up by the above process, mobile phone operators would reduce their profit lost substantially.
Appears in Collections:	[Department of Information Management & Graduate Institute of Information Management] Thesis

Files in This Item:

There are no files associated with this item.

View Licence

Loading...