MEMBERS

中心成員
  • 姓名 :

    徐立群

  • 職稱 :

    教授 Professor

  • 所屬實驗室 :

    金融預測分析與理財實驗室

  • 系所名稱 :

    成功大學財務金融研究所

  • 個人資料網頁 :

    點此進入

  • 學術成大 :

    點此進入

  • 洽談合作信箱 :

    shulc@mail.ncku.edu.tw

關於

研究領域

大數據預測分析、Fintech/舞弊偵測、機器學習、精準行銷、文字探勘、Chatbot、數位轉型

 

近期計畫

  • 應用應用資料探勘/機器學習於億進寢具實現精準行銷之研究 (產學)
  • 在大型路網中對不確定資料進行機率式CkNN查詢 (科技部)
  • 在隨選廣播環境下運用動態優先權排程來執行週期式查詢 (科技部)
  • 高效網路收集客資系統研發計畫(中*公司)
近期開課

〔開課名稱1〕:金融科技

〔開課系所1〕:會研所、財金所

〔課綱連結1〕:http://140.116.51.3/chinese/faculty/shulc/courses/Fintech/syllabus.htm


〔開課名稱2〕:資料探勘與大數據

〔開課系所2〕:會研所、財金所

〔課綱連結2〕:http://140.116.51.3/chinese/faculty/shulc/courses/dmbd/syllabus.htm


〔開課名稱3〕:電腦審計與安全

〔開課系所3〕:會研所

〔課綱連結3〕:http://140.116.51.3/chinese/faculty/shulc/courses/cas/syllabus.htm

近五年著作
  1. 〈金融科技(Fintech)如何重塑並創新金融服務?〉,【HBR導師講座】哈佛商業評論全球繁體中文版,2021年1月號。
  2. Using Deep Learning Techniques to Predict 10-Year US Treasury Yield. To appear in Proceedings of The 11th International Conference on Information Science and Technology (ICIST 2021), May 21 to May 23, 2021, Chendu, Sichuan, China.
  3. Shadow: Answering Why-not Questions On Top-k Spatial Keyword Queries Over Moving Objects. In Proceedings of The 26th International Conference on Database Systems for Advanced Applications (DASFAA 2021), April 11 to April 14 2021, Taipei, Taiwan. (Acceptance rate: 20%)
  4. LihChyun Shu, Bo Hsiao, Yi-Hsin Liou and Yi-Lin Tsai. On the Selection of Features for the Prediction Model of Cultivation of Sweet Potatoes at Early Growth Stage. IEEE Symposium Series on Computational Intelligence (IEEE SSCI 2019), Dec. 6 to Dec. 9 2019, Xiamen, China.
  5. Meifan Yang, LihChyun Shuand YuChih Wang. Mining Financial News to Predict S&P 500 Index Trend. International Conference on Innovation and Management (IAM 2019), July. 9 to July 12 2019, Hiroshima, Japan.
  6. Bo Hsiao, LihChyun Shu*, Fang-Yu Chou. Assessing the efficiency of the accounting industry using multiactivity network DEA: evidence from Taiwan. International Transactions in Operational Research 26(6): 2362-2386, 2019.
  7. Yanhong Li, Wang Zhang, Rongbo Zhu, Guohui Li, Maode Ma, LihChyun Shu, Changyin Luo. Fog-Based Pub/Sub Index With Boolean Expressions in the Internet of Industrial Vehicles. IEEE Transactions on Industrial Informatics 15(3): 1629-1642, 2019.
  8. Bo Hsiao, LihChyun Shu*, Ming-Miin Yu, Li-Kang Shen, Ding-Jiun Wang (2017, Nov). Performance evaluation of the Taiwan railway administration. Annals of Operations Research, 259(1-2): 119-156.
  9. Quan Zhou, Guohui Li, Jianjun Li, LihChyun Shu, Cong Zhang, Fumin Yang (2017, Jul). Dynamic priority scheduling of periodic queries in on-demand data dissemination systems. Information Systems, 67 p.58-70. (SCI). MOST 105- 2221-E-006-172.
  10. Yanhong Li, LihChyun Shu, Rongbo Zhu, Li Li (2017, May). A novel distributed air index for efficient spatial query processing in road sensor networks on the air.. Int. J. Communication Systems, 30(5) .
  11. Bo Hsiao, LihChyun Shu, Ti Chun Yeh (2017, Jul). Establishing a Stable Prediction Model of Loyal Customers for Repurchase Behavior. 21st Pacific Asia Conference on Information Systems, PACIS 2017, Langkawi, Malaysia. MOST 105-2221-E-006-172.
  12. Hairong Yu, Guohui Li, Jianjun Li, LihChyun Shu* (2016, Aug). DOcyclical: A Latency-Resistant Cyclic Multi-Threading Approach for Automatic Program Parallelization.. The Computer Journal, 59(8): 1155-1173.
  13. GuoQin Ning, Hongya Wang, LihChyun Shu, GuangRew Yeh (2016, May). Towards load shedding and scheduling schemes for data streams that maintain quality and timing requirements of query results. Soft Computing, 20(5): 1961- 1976.
  14. Hairong Yu, Guohui Li, LihChyun Shu*(2016, Feb). DSspirit: a data dependence and stride reference patterns profiling infrastructure.. The Journal of Supercomputing, 72(2): 770-788.
  15. Bo Hsiao, LihChyun Shu, Michal Young, Hsueh-Ching Yang (2016, Jul). Using Process Mining to Measure the Expected Costs of Business Processes. 20th Pacific Asia Conference on Information Systems, PACIS 2016, Chiayi, Taiwan, June 27 - July 1, 2016., 嘉義.
  16. Yanhong Li, Guohui Li, LihChyun Shu, Qun Huang, Hong Jiang (2015, Nov). Continuous Monitoring of Top- k Spatial Keyword Queries in Road Networks. Journal of Information Science and Engineering, 31(6): 1831-1848.
  17. Bo Hsiao, LihChyun Shu, Shu-Jeng Hsieh (2015, Aug). Predicting and testing probability of continuous purchasing of online customers.. Fuzzy Systems and Knowledge Discovery (FSKD), 2015 12th International Conference on.
  18. Bo Hsiao, Lihchyun Shu, Pin-Yeh Chen, Shu-Jeng Hsieh (2015, Jul). Trial Plan with Capitation Payment of the National Healthcare Insurance in Taiwan: Establishing a Royal Patient Selection Model. The 19th Pacific Asia Conference on Information Systems (PACIS 2015), 新加坡.

 

產學成果

1.億進寢具生活館

精準行銷 — 億進寢具生活館一年四季經常有行銷活動,它有24萬會員,但每次行銷活動僅挑選部分會員,期望能達到最高的行銷效果。我們協助億進寢具,從2018年8月起提供精準行銷預測名單。其中排序前6000名的精準行銷名單,較以往高出了3-4倍的回購率。

 

2.INSTO

交易舞弊偵測 — INSTO是一個行動收付款平台(https://ins.to/tw/),2013年成立於美國Kansas City,台灣INSTO服務於2017年上線。賣家可直接透過手機APP或網站,依照買賣雙方的需求,自訂客製化的分期/定期付款方案。INSTO的「一次付清」功能讓各式中小型賣家免去申請及維護實體刷卡機的煩惱,直接使用INSTO APP做為行動刷卡機,買賣雙方交易更輕鬆方便。INSTO平台交易金額從2019第一季1千8百萬成長到第四季1億6千萬,成長了874%。

INSTO委託本實驗室研發其平台舞弊詐欺系統。INSTO目前交易違約率達6%,我們實驗室利用機器學習建立預測模型,經過數次模型改進,目前sensitivity可達84%,specificity達99%,刻正持續改進模型。預計不久模型表現將趨於穩定,屆時INSTO將利用我們的模型降低交易違約率,同時增加接受違約機率低的交易,提高服務費之收取,增加其營收。

Top