การพัฒนาระบบตรวจสอบการซื้อขายหลักทรัพย์ที่ผิดปกติ สำหรับบริษัทหลักทรัพย์ในประเทศไทยโดยการใช้ดาต้าไมนิ่ง

• Main Author: ภาณุ ตันโสรัจประเสริฐ
• Other Authors: ยรรยง เต็งอำนวย
• Format: Theses and Dissertations
• Language: Thai
• Published: จุฬาลงกรณ์มหาวิทยาลัย 2003
• Subjects: ตลาดหลักทรัพย์แห่งประเทศไทย; ดาต้าไมนิง; หลักทรัพย์ > กฎหมายและระเบียบข้อบังคับ > ไทย; ตลาดหลักทรัพย์
• Online Access: https://digiverse.chula.ac.th/Info/item/dc:45818

Nowadays, there are growing higher number of investments in Stock Exchange of Thailand. New stock trading regulations by Stock Exchange of Thailand has been announced imperatively to control and protect an abnormal stock trading for both of individual and institute investors. Therefore, broker penalty cost is also higher accordingly. In order to reduce cost of penalty, this research is conducting the development of audit system to facilitate auditor to monitor abnormal stock trading efficiently. Development tool for this research utilize Data Mining Clustering in Microsoft SQL server to analyze stock trading data. The process started from SET regulation announcements are converted to data mining rules. The next step is filtering and appending of Auto-T protocol that is stock trading standard format provided by Stock Exchange of Thailand into the database. Finally, data mining clustering tools will generate the valuable results from Auto-T information stored in the database. The result of this research represents the figure in term of clustering format. Each cluster displays the stock trading patterns having the same behavior. Each cluster is able to facilitate auditor to search a group of potential customer who send abnormal stock trading order easily. Moreover, data mining clustering tools also provide the probability percentage for auditor to specify abnormal level for proactive auditing actions in the future.