機器學習與會計科目判斷－台灣半導體公司使用人工智慧提升行政效率個案對會計教育之啟示

Machine Learning and Account Classification Judgements: Educational Implications of the Application of Artificial Intelligence to Accounting Procedures of a Taiwan Semiconductor Company

劉順仁(Shuen-Zen Liu) ； 王俊凱(Chun-Kai Wang) ； 李季澄(Chi-Cheng Lee) ； 黃禹翔(Yu-Shiang Huang)

《中華會計學刊》 17卷2期 (2021/12) Pp. 245-268

https://doi.org/10.6538/TAR.202112_17(2).0002

摘要

本文討論台灣某半導體公司（以下簡稱F公司）如何應用人工智慧技術建構自動判斷會計科目正確性的機器學習模型。本研究衍生自F公司財務處與某國立大學會計學系（以下簡稱A系）合作的「暑期人工智慧實習專案」，該專案期間為2019年7月1日到2019年8月31日。本文有兩個貢獻：（1）探討在人工智慧時代，會計教育界與產業界如何合作發展數據分析（data analytics）與產業實作課程（capstone course），以協助會計教育的數位轉型。（2）本文討論企業如何以機器學習技術，有效地降低成本並提升其會計作業行政流程效率。本研究期望能夠以此協助更多臺灣企業，更廣泛的運用機器學習技術來優化企業流程。

關鍵字

產業實作課程 ； 機器學習模型 ； 會計科目判斷

並列摘要

This study discusses how a Taiwan semiconductor company (F company hereafter) applies artificial intelligence (AI) techniques to build machine learning models for automatic account classification judgements. The study builds on a 2019 summer internship project collaborated by the finance department of F company and the department of accounting of a national university in Taiwan. The study contributes to the accounting literature in two ways. (1) The study provides implications for accounting educators on how to collaborate with industry practitioners in terms of conducting AI problem-solving projects in building data-driven capstone courses. Thus, the study helps expedite the digital transformation of accounting education. (2) Additionally, the study shows how a large company builds machine learning models to save personnel costs and enhance operational efficiency in its accounting procedures. The study aims to motivate Taiwanese companies to explore opportunities of optimizing their operational efficiency through AI techniques.

並列關鍵字

Capstone courses ； Machine learning models ； Account classification judgements

參考文獻

Ding, Kexing, Baruch Lev, Xuan Peng, Ting Sun, and Miklos A. Vasarhelyi. “Machine Learning Improves Accounting Estimates: Evidence from Insurance Payments.” Review of Accounting Studies 25, no. 3 (2020): 1098–1134. https://doi.org/10.1007/s11142-020-09546-9. EY. “How an AI Application Can Help Auditors Detect Fraud.” EY. EY, January 7, 2019. https://go.ey.com/2PphF1D.

Gers, Felix A. “Learning to Forget: Continual Prediction with LSTM.” 9th International Conference on Artificial Neural Networks: ICANN '99, 1999. https://doi.org/10.1049/cp:19991218. Googel Developers “How Machine Learning with TensorFlow Enabled Mobile Proof-of-Purchase at Coca-Cola.” Google Developers Blog, September 21, 2017. https://developers.googleblog.com/2017/09/how-machine-learning-with-tensorflow.html.

Guo, Li, Feng Shi, and Jun Tu. “Textual Analysis and Machine Leaning: Crack Unstructured Data in Finance and Accounting.” The Journal of Finance and Data Science 2, no. 3 (2016): 153–70. https://doi.org/10.1016/j.jfds.2017.02.001.

Li, Bin, Julia Yu, Jie Zhang, Bin Ke, Boa Yang. “Detecting Accounting Fraud in Publicly Traded U.S. Firms Using a Machine Learning Approach.” Journal of Accounting Research 58, no. 1 (2020): 199–235. https://doi.org/10.1111/1475-679x.12292.

Li, Yachao, and Yufa Wang. “Machine Learning Methods of Bankruptcy Prediction Using Accounting Ratios.” Open Journal of Business and Management 06, no. 01 (2018): 1–20. https://doi.org/10.4236/ojbm.2018.61001.