Loading...
|
Please use this identifier to cite or link to this item:
https://nccur.lib.nccu.edu.tw/handle/140.119/159700
|
| Title: | 以電話會議紀錄文本建構營運指標：流程與實證分析
Developing Operational Performance Indicators from Earnings Conference Calls: Process and Empirical Analysis |
| Authors: | 黃念祺
Huang, Nian-Qi |
| Contributors: | 莊皓鈞
周彥君
Chuang, Hao-Chun
Chou, Yen-Chun
黃念祺
Huang, Nian-Qi |
| Keywords: | 電話會議紀錄文本
營運績效指標
SBERT
FinBERT
Earnings conference calls
Operational performance indicators
SBERT
FinBERT |
| Date: | 2025 |
| Issue Date: | 2025-10-02 11:10:53 (UTC+8) |
| Abstract: | 企業電話會議紀錄逐字稿蕴含豐富的非結構化資訊，能揭示管理層對營運績效與供應鏈策略的洞察，彌補傳統財務指標僅能量化過去成果的不足。然而，既有研究鮮少系統性地從中建構能反映企業內部營運狀況的指標，且早期的自然語言處理 (Natural Language Process, NLP) 方法在捕捉複雜語意上仍有其限制。
為此，本研究提出一套比較性的文本分析框架，旨在評估兩種不同路徑在建構多維度營運績效指標上的有效性。第一種是數據驅動的「字典法」，透過機器學習自動從歷史與同業文本中挖掘詞彙，並動態生成「種子句」；第二種是專家知識導向的「查詢法」，採用一組預先定義的負面情境「查詢句」作為語意錨點。兩種方法確立輸入句後，皆運用 Sentence-BERT (SBERT) 的上下文語意嵌入技術，針對「供應商」(supplier)、「顧客」(customer)、「存貨」(inventory) 與「風險」(risk) 四大核心營運構面，識別出關鍵討論句，並透過 FinBERT 模型將其量化為情感指標。
本研究的貢獻在於，透過比較兩種不同設計的方法論，共同驗證了從文本敘事中提取的營運指標，確實與企業財務績效存在顯著統計關聯。此框架不僅補充了傳統財務分析的不足，也為供應鏈管理的實證研究提供了一個穩健、可複製的分析途徑。
Earnings conference call transcripts offer rich insights into operational performance, complementing traditional financial metrics. However, prior research has rarely constructed systematic operational indicators from these texts, and early Natural Language Processing (NLP) methods struggle with complex semantics.
This study proposes a comparative framework evaluating two approaches: a data-driven "dictionary method" that generates dynamic "seed sentences," and an expert-driven "query method" using predefined negative "query sentences" as semantic anchors. Both approaches utilize Sentence-BERT (SBERT) to identify key discussions across four dimensions—supplier, customer, inventory, and risk—and then quantify them into sentiment indicators using the FinBERT model.
Our contribution lies in validating, through this dual-method comparison, that textual indicators have a significant statistical association with firm financial performance. This framework complements traditional analysis and provides a robust, replicable pathway for empirical research in supply chain management. This framework complements traditional analysis and provides a robust, replicable pathway for empirical research in supply chain management. |
| Reference: | Araci, D. (2019). Finbert: Financial sentiment analysis with pre-trained language models. arXiv preprint arXiv:1908.10063.
Bolstorff, P., & Rosenbaum, R. G. (2007). Supply chain excellence: a handbook for dramatic improvement using the SCOR model. AMACOM/American Management Association.
Chopra, S., & Meindl, P. (2007). Supply chain management. Strategy, planning & operation (pp. 265-275). Gabler.
Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. (2019, June). Bert: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies, volume 1 (long and short papers) (pp. 4171-4186).
Ersahin, N., Giannetti, M., & Huang, R. (2024). Supply chain risk: Changes in supplier composition and vertical integration. Journal of International Economics, 147, 103854.
Fu, X., Wu, X., & Zhang, Z. (2021). The information role of earnings conference call tone: Evidence from stock price crash risk. Journal of Business Ethics, 173, 643-660.
Huang, A. H., Wang, H., & Yang, Y. (2023). FinBERT: A large language model for extracting information from financial text. Contemporary Accounting Research, 40(2), 806-841.
Kesavan, S., & Mani, V. (2013). The relationship between abnormal inventory growth and future earnings for U.S. public retailers. Manufacturing & Service Operations Management, 15(1), 6–23.
Kesavan, S., Gaur V., Raman A. (2010). Do inventory and gross margin data improve sales forecasts for US public retailers? Management Science, 56(9): 1519–1533.
Li, F. (2010). The information content of forward‐looking statements in corporate filings—A naïve Bayesian machine learning approach. Journal of accounting research, 48(5), 1049-1102.
Loughran, T., & McDonald, B. (2011). When is a liability not a liability? Textual analysis, dictionaries, and 10‐Ks. The Journal of finance, 66(1), 35-65.
Maaten, L. V. D., & Hinton, G. (2008). Visualizing data using t-SNE. Journal of machine learning research, 9(Nov), 2579-2605.
Matsumoto, D., Pronk, M., & Roelofsen, E. (2011). What makes conference calls useful? The information content of managers' presentations and analysts' discussion sessions. The Accounting Review, 86(4), 1383-1414.
Mikolov, T., Chen, K., Corrado, G., & Dean, J. (2013). Efficient estimation of word representations in vector space. arXiv preprint arXiv:1301.3781.
Price, S. M., Doran, J. S., Peterson, D. R., & Bliss, B. A. (2012). Earnings conference calls and stock returns: The incremental informativeness of textual tone. Journal of Banking & Finance, 36(4), 992-1011.
Reimers, N., & Gurevych, I. (2019). Sentence-bert: Sentence embeddings using siamese bert-networks. arXiv preprint arXiv:1908.10084.
Sanh, V., Debut, L., Chaumond, J., & Wolf, T. (2019). DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter. arXiv preprint arXiv:1910.01108.
Singhal, A., Buckley, C., & Mitra, M. (2017). Pivoted document length normalization ACM SIGIR Forum. ACM, New York, NY, USA, 176-184.
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., ... & Polosukhin, I. (2017). Attention is all you need. Advances in neural information processing systems, 30.
Wu, D. (2024). Text-based measure of supply chain risk exposure. Management Science, 70(7), 4781-4801. |
| Description: | 碩士
國立政治大學
資訊管理學系
112356019 |
| Source URI: | http://thesis.lib.nccu.edu.tw/record/#G0112356019 |
| Data Type: | thesis |
| Appears in Collections: | [資訊管理學系] 學位論文
|
Files in This Item:
| File |
Size | Format | |
|---|
| 601901.pdf | 884Kb | Adobe PDF | 0 | View/Open |
|
All items in 政大典藏 are protected by copyright, with all rights reserved.
|
