進入障礙是聯邦法院在分析反托拉斯案件時，使用的一個重要概念。此外，進入障礙概念的使用頻率，可能有助於反托拉斯律師考慮該概念是否應該成為給定案件的中心論點。儘管如此，很少有文獻透過更合乎邏輯的方法，討論進入障礙的概念如何運用於反托拉斯的案件。為了研究這個問題，筆者使用 Latent Dirichlet Allocation (LDA)主題模型來分析從1910年以來在上下文中使用進入障礙詞彙的2208起反托拉斯案件。因此，本研究發現聯邦法院很可能在六個不同的主題中使用進入障礙的詞彙：航空業的合併與收購、一般價格操縱案件、醫療服務產業的壟斷、專利許可協議、產品搭售安排、存在使用權問題的網路服務營運商（例如，Comcast）。 此外，如果逐年調查這六個類別，本研究的結果會揭示以下內容。
首先，雖然反托拉斯法是在19世紀末通過的，但進入障礙的概念直到1970年才被聯邦法院廣泛採納。其次，進入障礙的詞彙在專利許可協議、產品搭售安排等反托拉斯案件中被越來越多地使用。另一方面，只有少數醫療服務產業的壟斷案件會使用進入障礙的詞彙。第三，聯邦法院在1970年至1990年的航空業合併案中經常使用進入障礙的詞彙，但後來聯邦法院很少使用這個詞彙。這種趨勢類似於網路服務營運商使用權問題的主題，在2005年之前的案件中也普遍使用進入障礙的詞彙。有了這個分析，不僅可以幫助反托拉斯律師在未來的反托拉斯案件中加強法律論證，而且還可以預測哪種類型的行業可能更重視進入障礙的概念。

Entry barrier is an important concept that has been used by the Federal Court in analyzing the antitrust cases. Moreover, the frequency of the usage of the entry barrier concept might be helpful for antitrust lawyers to consider whether that concept should be the center argument in the given case. Nonetheless, little literature has discussed how the concept of the entry barrier is likely to be used in a Federal antitrust case through a more logical method. To investigate this issue, we use Latent Dirichlet Allocation (LDA) topic modeling to analyze 2208 antitrust cases with entry barrier wordings in the context since 1910. As a result, we found that the Federal Courts are likely to use entry barrier wordings in six different topics: merger and acquisition in airline industry, general price-fixing cases, monopolization in hospital service industry, patent licensing agreement, product tying arrangement, the internet service carrier with issues of access (e.g., Comcast). Furthermore, if we investigate those six categories year by year, our results reveal following things. First, although the antitrust law was passed in late 19 th century, the concept of entry barrier has not been widely adopted by the Federal Court until 1970. Second, the concept of the entry barrier has increasingly been used in the antitrust cases about the patent licensing agreement and the product tying arrangement. On the other hand, only few of the monopolization in hospital service industry cases would use entry barrier concept.
Third, the Federal Court used the concept of the entry barrier very often in the context of the merger in the airline industry from 1970 to 1990, but later the Federal Court used the concept less regularly. This trend is similar to the topic about the internet service carrier with issues of access, which has popularly used entry concept in the cases until 2005. Having this analysis could help antitrust lawyers not only strengthen their legal arguments in the future antitrust cases, but also predict what type of industries might take the entry barrier concept more seriously.

論文審定書 i
誌謝 ii
摘要 iii
Abstract iv
目 錄 v
圖 次 vi
表 次 vii
壹、前言 1
一、研究動機 1
二、研究背景 2
三、研究問題 5
貳、文獻回顧 6
一、自然語言處理的技術和應用 6
二、進入障礙觀念之演變 8
三、進入障礙與美國反托拉斯判決 10
四、進入障礙詞彙在美國反托拉斯判決的使用 12
參、研究方法 14
一、資料來源及處理 14
二、研究變數介紹 19
三、研究分析工具 22
肆、研究結果 32
一、敘述性統計 32
二、主題模型分析 39
伍、研究結論與建議 50
一、研究結論 50
二、研究限制與未來發展 51
參考文獻 53

Agarwal, B., Mittal, N., Bansal, P., & Garg, S. (2015). Sentiment Analysis Using Common-Sense and Context Information. Journal of Computational Intelligence and Neuroscience., 2015, 1028-1036.
Bain, J. S. (1956). Barriers to new competition. Cambridge, MA: Harvard University Press.
Blei, D. M., Ng, A. Y., & Jordan, M. I. (2003). Latent dirichlet allocation. J. Mach. Learn. Res., 3(2), 993-1022.
Blei, D. M. (2012). Probabilistic topic models. Communications of the ACM, 55(4), 77-84.
Bui, Q.V., Sayadi, K., Bui, M. (2015). A multi-criteria document clustering method based on topic modeling and pseudoclosure function. In: Proceedings of the sixth international symposium on information and communication technology, 38-45.
Cao, S. (2022). Economic analysis in antitrust law: An automated approach applied to us appellate courts. Stanford Computational Antitrust, 2(8), 156-176.
Carlton, D. W. (2008). Barriers to entry. Issues in competition law and policy, 1, 601-617.
Chen, T. C., & Hsu, C. H. (2020) China’s human rights foreign policy in the Xi Jinping era: Normative revisionism shrouded in discursive moderation. The British Journal of Politics and International Relations, 23(2), 1-20.
Christian, H., Agus, M.P., Suhartono, D. (2016). Summarization using term frequency inverse document frequency (TF-IDF). Com. Tech. 7(4), 285-294.
Darling W. M. (2011). A theoretical and practical implementation tutorial on topic modeling and gibbs sampling. In Proceedings of the 49th annual meeting of the association for computational linguistics: Human language technologies, 642-647.
Demsetz, H. (1982). Barriers to entry. American Economic Review, American Economic Association, 72(1), 47-57.
Dreisbach, C., Koleck, T. A., Bourne, P. E., & Bakken, S. (2019). A systematic review of natural language processing and text mining of symptoms from electronic patient authored text data. International Journal of Medical Informatics, 125, 37-46.
Fang, X., Zhan, J. (2015). Sentiment analysis using product review data. Journal of Big Data, 2(5), 1-14.
Ferguson, J. M. (1974). Advertising and competition: Theory, measurement, fact. Cambridge, MA: Ballinger.
Griffiths, T. L. & Steyvers, M. (2004). Finding scientific topics. Proceedings of the National academy of Sciences of the United States of America, 101(1), 5228-5235.
Gupta, V., & Lehal, G. (2009). A survey of text mining techniques and applications. Journal of Emerging Technologies, 1(1), 60-76.
Hong, L., Martinez, E. F. & Martinez, V. F. (2016) Topic Models to Infer Socio-Economic Maps. Proceedings of the AAAI Conference on Artificial Intelligence, 30(1), 3835-3841.
Hu, M. and Liu, B. (2005). Mining and summarizing customer reviews. Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing.
Jelodar, H., Wang, Y., Yuan, C., Feng, X., Jiang, X., Li, Y., & Zhao, L. (2019). Latent dirichlet allocation (LDA) and topic modeling: models, applications, a survey. Multimedia Tools and Applications, 78, 15169-15211.
Liddy, E.D. 2001. Natural Language Processing. In M. A. Drake (Ed.), Encyclopedia of Library and Information Science (2nd ed.). New York: Marcel Decker, Inc.
McAFee, P. R., Mialon, H. M. & Williams, M. A. (2004). What is a barrier to entry?. American Economic Review, 94 (2), 461-465.
Mumcuoglu, E., Ozturk, C. E., Ozaktas, H. M., & Koc, A. (2021). Natural language processing in law: Prediction of outcomes in the higher courts of Turkey. Information Processing & Management, 58(5),102684.
Naseem, U., Razzak, I., Khan, S. K., Prasad, M. (2020). A Comprehensive Survey on Word Representation Models: From Classical to State-Of-The-Art Word Representation Language Models. ACM Transactions on Asian and Low-Resource Language Information Processing. arXiv preprint arXiv:2010.15036.
Schmalensee, R. (2004). Sunk costs and antitrust barriers to entry. American Economic Review, 94 (2), 471-475.
Schrepel, T. (2021). Computational antitrust: An introduction and research agenda. Stanford Computational Antitrust, 1(1), 1-15.
Steyvers, M & Griffiths, T. (2007). Probabilistic topic models. Handbook of latent semantic analysis, 427(7), 424-440.
Stigler, G. J. (1968). The organization of industry. Chicago, IL: University of Chicago Press.
Sun, S., Luo, C., & Chen, J. (2017). A review of natural language processing techniques for opinion mining systems. Information fusion, 36, 10-25.
von Weizsacker, C. C. (1980). A welfare analysis of barriers to entry. Bell Journal of Economics, 11(2), 399-420.