有鑒於網路市場的蓬勃發展，使得線上購物的消費人口大量增加，網路行銷模式亦成為企業主要的行銷工具，然而大量的網路行銷導致商標圖樣更廣泛的被使用，也使商標侵權發生的頻率提高。Facebook、Line等社群媒體的普及，使用通訊軟體聊天成為目前人與人之間最常見的互動方式，造就聊天機器人這種新的服務型態。聊天機器人（Chatbot）是經由對話或文字進行交談的應用程式，藉由自然語言學習與和資訊檢索模擬人類對話，給予適當的回應。當面臨商標相關問題時，大眾常缺少了快速獲取答案的管道，因此本研究以商標常見問答FAQ作為聊天機器人之任務並進行系統設計和開發，以聊天機器人作為介面，讓人們可以透過簡短的問句理解商標註冊和侵權等相關知識並獲得需要的答案。
研究主要分為兩個部分，建構商標FAQ本體論及聊天機器人建構，本體論作為聊天機器人之知識背景、對話邏輯並支撐驅動程式運行。本研究根據台灣商標法建構本體論，分為意圖、子意圖及子意圖關鍵字三個階層。首先透過台灣智財局網站商標常見問題分類將本體論分為14個意圖，並透過商標法將範圍較大之意圖向下延伸至數個子意圖。接著藉由Term Frequency - Inverse Document Frequency找出子意圖的關鍵詞彙作為本體論之第三層，本體論中每一個子意圖皆含有一個可以回應的答案。完成本體論的建構後將其轉換為JSON資料模式，作為聊天機器人的資料庫。聊天機器人之檢索根據本體論分為兩個階段，分別為問題分類和關鍵字分析，在問題分類中，先透過RNN-LSTM訓練，獲得問題之意圖後並進行資料庫的檢索，若意圖中沒有子意圖便直接生成回答；反之則進行第二階段，將問句與子意圖下的關鍵字做相似度分析。本研究使用word2vec生成詞向量，計算詞彙之間的餘弦相似度(cosine similarity)得出平均分數最高的子意圖，根據最後得到之意圖查詢資料庫生成回應之答案。最後，本研究透過Django網頁架構建立網頁應用程式，目的是希望本研究之結果可以作為實際上之應用。

Recently, enterprises in Taiwan began to develop their own brands and trademarks so the importance of trademark increase rapidly. The development of the Internet led to more trademark (TM) infringement cases over the Internet, which makes the brand protection more difficult. When people encounter questions related to TM registration, infringement and other issues, which may be hard to find the correct answers. Therefore, this study builds a chatbot platform to provide users with simple, easy, and quick answers to their questions. This study is mainly devided in two part, first one is to build the ontology of trademark consultation based on Taiwan IP Office (TIPO) TM frequent asked questions (FAQs), which aims to mapping the chatbot process and provide for knowledge database. The second is to construct the chatbot system that embedded with natural language processing capabilities.
The TM consultation ontology is built on hierarchical structure with three levels, namely intent, sub-intent and key terms of sub-intent. Intent and sub-intent are defined according to the classifications of FAQ in TIPO and literature of TM law. For the third level, we use term frequency - inverse document frequency method to extract the key terms. The process of retrieval system includes intent prediction and key term similarity analysis. In intent prediction, we use long short-term memory model from recurrent neural network to train the classifier. For similarity analysis, it generated word vectors by word2vec model, and then calculated the cosine similarity between key terms and question to obtain the sub-intent with the highest average score and response answers by querying the database. Last, this study builds a web application through Django web architecture in order to make this study available for practical use.

摘要 I
ABSTRACT II
目錄 III
圖目錄 V
表目錄 VI
壹、緒論 1
1.1 研究背景與動機 1
1.2 研究範圍與目的 2
1.3 研究方法與步驟 3
貳、文獻探討 5
2.1 數位行銷與商標侵權 5
2.1.1 商標 5
2.1.2 商標侵權 6
2.1.3 網路行銷中的侵權行為 7
2.2 本體論 10
2.2.1 本體論定義 10
2.2.2 本體論工程 12
2.3聊天機器人 14
2.3.1 定義與發展 14
2.3.2 問答系統 17
2.3.3 基於本體論之問答系統 18
2.3.4 對話生成模型 20
2.4 自然語言處理 22
2.4.1 Word Embedding 22
2.4.2. 問題分類 23
2.4.3. 循環神經網絡模型 25
參、研究方法 28
3.1 知識本體論與資料建構 30
3.1.1 商標問答相關資料蒐集 31
3.1.2 建構本體論 32
3.1.3 關鍵字擷取 32
3.1.4 資料庫建立 34
3.2諮詢機器人對話系統 36
3.2.1 意圖預測 37
3.2.2 相似度分析 41
肆、案例分析 44
4.1 知識本體論建立 44
4.1.1. 資料蒐集 44
4.1.2. 建立本體論 44
4.1.3. 萃取子意圖關鍵字 (Key terms) 46
4.1.4. 整理成JSON資料庫 47
4.2 諮詢機器人對話系統 48
4.2.1 意圖預測 48
4.2.2 相似度分析 50
4.2.3 答案生成 51
4.2.4 系統評估 52
伍、結論和建議 55
5.1 結論 55
5.2 建議 56
REFERENCES 57
APPENDIX A. 商標諮詢本體論資料表 66
APPENDIX B. 商標諮詢回應 68
APPENDIX C. 商標問題訓練集 73
APPENDIX D. 系統評估結果 93
APPENDIX E. 口語測試問句 96

[1] Allam, A. M. N., & Haggag, M. H. (2012). The question answering systems: A survey. International Journal of Research and Reviews in Information Sciences (IJRRIS), 2(3).
[2] Al-Zubaide, H., & Issa, A. A. (2011, November). Ontbot: Ontology based chatbot. In Innovation in Information & Communication Technology (ISIICT), 2011 Fourth International Symposium on (pp. 7-12). IEEE.
[3] Athira, P. M., Sreeja, M., & Reghuraj, P. C. (2013). Architecture of an ontology-based domain-specific natural language question answering system. International Journal of Web & Semantic Technology, 4(4), 31.
[4] Bradeško, L., & Mladenić, D. (2012). A survey of chatbot systems through a loebner prize competition. In Proceedings of Slovenian Language Technologies Society Eighth Conference of Language Technologies (pp. 34-37).
[5] Berners, L.T., Handler, J. and Lassila, O., 2001, “The Semantic Web,” Scientific American, vol. 184.
[6] Bunge, M. (1977). “Treatise on basic philosophy: Ontology I: the furniture of the world,” Vol. 1, Springer Science & Business Media.
[7] Berg, M., & Düsterhöft, A. (2010). Website Interaction with Text-based Natural Language Dialog Systems. 7. Wismarer Wirtschaftsinformatiktage.
[8] Bhargava, A., Celikyilmaz, A., Hakkani-Tür, D., & Sarikaya, R. (2013, May). Easy contextual intent prediction and slot detection. In Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International Conference on (pp. 8337-8341). IEEE.
[9] Beckett, D., 2014, “RDF 1.1 XML Syntax,” W3C Recommendation, available at: http://www.w3.org/TR/rdf-syntax-grammar/
[10] Berant, J., & Liang, P. (2014). Semantic parsing via paraphrasing. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (Vol. 1, pp. 1415-1425).
[11] Brookfield Communications, Inc. v. West Coast Entertainment Corp., 174 F.3d 1036, 1044 (9th Cir. 1999)
[12] Bordes, A., Chopra, S., & Weston, J. (2014). Question answering with subgraph embeddings. arXiv preprint arXiv:1406.3676.
[13] Bouziane, A., Bouchiha, D., Doumi, N., & Malki, M. (2015). Question answering systems: survey and trends. Procedia Computer Science, 73, 366-375.
[14] Bu, F., Zhu, X., Hao, Y., & Zhu, X. (2010, October). Function-based question classification for general QA. In Proceedings of the 2010 conference on empirical methods in natural language processing (pp. 1119-1128). Association for Computational Linguistics.
[15] Chasseur, C., Li, Y., & Patel, J. M. (2013, June). Enabling JSON Document Stores in Relational Systems. In WebDB (Vol. 13, pp. 14-15).
[16] Cox, S. (2003). Discriminative techniques in call routing. Proceedings of ICASSP, Hong Kong, April 2003.
[17] Curé, O., Kerdjoudj, F., Faye, D., Le Duc, C., & Lamolle, M. (2013). On the potential integration of an ontology-based data access approach in NoSQL stores. International Journal of Distributed Systems and Technologies (IJDST), 4(3), 17-30.
[18] Eid, R., & El-Gohary, H. (2013). The impact of E-marketing use on small business enterprises' marketing success. The Service Industries Journal, 33(1), 31-50.
[19] Fader, A., Zettlemoyer, L., & Etzioni, O. (2014, August). Open question answering over curated and extracted knowledge bases. In Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining (pp. 1156-1165). ACM.
[20] Financial Times Lexicon. (2015). “Definition of digital marketing,” available at: http://lexicon.ft.com/Term?term=digital-marketing
[21] Frank, A., Krieger, H. U., Xu, F., Uszkoreit, H., Crysmann, B., Jörg, B., & Schäfer, U. (2007). Question answering from structured knowledge sources. Journal of Applied Logic, 5(1), 20-48.
[22] Ferrucci, D., Levas, A., Bagchi, S., Gondek, D., & Mueller, E. T. (2013). Watson: beyond jeopardy!. Artificial Intelligence, 199, 93-105.
[23] Goldman, E. (2008). Online word of mouth and its implications for trademark law. In G.B. Dinwoodie & M.D. Janis (Eds.), Trademark law and theory a handbook of contemporary research (pp. 404–429). Cheltenham, UK: Edward Elgar.
[24] Grau, S., Sanchis, E., Castro, M. J., & Vilar, D. (2004). Dialogue act classification using a Bayesian approach. In 9th Conference Speech and Computer.
[25] Gruber, T. (2008). Ontology. Entry in the Encyclopedia of Database Systems, Ling Liu and M. Tamer Özsu.
[26] Gupta, P., & Gupta, V. (2012). A survey of text question answering techniques. International Journal of Computer Applications, 53(4).
[27] Ittycheriah, A., Franz, M., Zhu, W. J., Ratnaparkhi, A., & Mammone, R. J. (2000, November). IBM's Statistical Question Answering System. In TREC.
[28] Johnson, R., & Zhang, T. (2016). Supervised and semi-supervised text categorization using LSTM for region embeddings. arXiv preprint arXiv:1602.02373.
[29] Kilejian, M., & Dahlstrom, S. (2016). Trademark Infringement Claims in Keyword Advertising. Franchise Law Journal, 36(1).
[30] Kim, Y. (2014). Convolutional neural networks for sentence classification. arXiv preprint arXiv:1408.5882.
[31] Kim, S. N., Medelyan, O., Kan, M. Y., & Baldwin, T. (2010, July). Semeval-2010 task 5: Automatic keyphrase extraction from scientific articles. In Proceedings of the 5th International Workshop on Semantic Evaluation (pp. 21-26). Association for Computational Linguistics.
[32] Levy, O., & Goldberg, Y. (2014). Neural word embedding as implicit matrix factorization. In Advances in neural information processing systems (pp. 2177-2185).
[33] Liu, B. (2012). Sentiment analysis and opinion mining. Synthesis lectures on human language technologies, 5(1), 1-167.
[34] Lopez, V., Uren, V., Sabou, M., & Motta, E. (2011). Is question answering fit for the semantic web?: a survey. Semantic Web, 2(2), 125-155.
[35] Mikolov, T., Sutskever, I., Chen, K., Corrado, G. S., & Dean, J. (2013). Distributed representations of words and phrases and their compositionality. In Advances in neural information processing systems (pp. 3111-3119).
[36] Minaee, S., & Liu, Z. (2017). Automatic question-answering using a deep similarity neural network. arXiv preprint arXiv:1708.01713.
[37] Mishra, A., & Jain, S. K. (2016). A survey on question answering systems with classification. Journal of King Saud University-Computer and Information Sciences, 28(3), 345-361.
[38] Moniruzzaman, A. B. M., & Hossain, S. A. (2013). Nosql database: New era of databases for big data analytics-classification, characteristics and comparison. arXiv preprint arXiv:1307.0191.
[39] Nio, L., Sakti, S., Neubig, G., Toda, T., Adriani, M., & Nakamura, S. (2014). Developing non-goal dialog system based on examples of drama television. In Natural Interaction with Robots, Knowbots and Smartphones (pp. 355-361). Springer, New York, NY.
[40] North American Medical, Corp. v. Axiom Worldwide, Inc. No. 07-11574 (11th Cir., April 7, 2008)
[41] Polaroid Corp. v. Polarad Elect. Corp., 287 F.2d 492 (1961).
[42] Ramesh, K., Ravishankaran, S., Joshi, A., & Chandrasekaran, K. (2017, May). A Survey of Design Techniques for Conversational Agents. In International Conference on Information, Communication and Computing Technology (pp. 336-350). Springer, Singapore.
[43] Ravuri, S., & Stolcke, A. (2015). Recurrent neural network and lstm models for lexical utterance classification. In Sixteenth Annual Conference of the International Speech Communication Association.
[44] Rescuecom Corp. v. Google, Inc. 456 F. Supp. 2d 393 (N.D.N.Y. 2006)
[45] Rong, X. (2014). word2vec parameter learning explained. arXiv preprint arXiv:1411.2738.
[46] Serban, I. V., Lowe, R., Henderson, P., Charlin, L., & Pineau, J. (2015). A survey of available corpora for building data-driven dialogue systems. arXiv preprint arXiv:1512.05742.
[47] Serban, I. V., Sordoni, A., Bengio, Y., Courville, A. C., & Pineau, J. (2016, February). Building End-To-End Dialogue Systems Using Generative Hierarchical Neural Network Models. In AAAI (pp. 3776-3784).
[48] Scott, C. D. (2013). Trademark strategy in the internet age: Customer hijacking and the doctrine of initial interest confusion. Journal of Retailing, 89(2), 176-189.
[49] Shen, D., & Lapata, M. (2007). Using semantic roles to improve question answering. In Proceedings of the 2007 joint conference on empirical methods in natural language processing and computational natural language learning (EMNLP-CoNLL).
[50] Shi, Y., Yao, K., Chen, H., Pan, Y. C., Hwang, M. Y., & Peng, B. (2015, April). Contextual spoken language understanding using recurrent neural networks. In Acoustics, Speech and Signal Processing (ICASSP), 2015 IEEE International Conference on (pp. 5271-5275). IEEE.
[51] Sokolova, M., & Lapalme, G. (2009). A systematic analysis of performance measures for classification tasks. Information Processing & Management, 45(4), 427-437.
[52] Stehman, S. V. (1997). Selecting and interpreting measures of thematic classification accuracy. Remote sensing of Environment, 62(1), 77-89.
[53] Studer, R., Benjamins, V. R., Fensel, D., 1998, “Knowledge Engineering: Principles and Methods”, Data and knowledge engineering, Vol. 25, pp. 161-197.
[54] Suresh kumar, G., Zayaraz, G., 2014. Concept relation extraction using Naıve Bayes classifier for ontology-based question answering systems. J. King Saud Univ
[55] Swartout, W., & Tate, A. (1999). Ontologies. IEEE Intelligent Systems and Their Applications, 14(1), 18-19.
[56] Tiago, M. T. P. M. B., & Veríssimo, J. M. C. (2014). Digital marketing and social media: Why bother?. Business Horizons, 57(6), 703-708.
[57] Toptsis, I., Li, S., Wrede, B., & Fink, G. A. (2004). A multi-modal dialog system for a mobile robot. In Eighth International Conference on Spoken Language Processing.
[58] Turing, A. M. (2009). Computing machinery and intelligence. In Parsing the Turing Test (pp. 23-65). Springer, Dordrecht.
[59] Uschold, M., & King, M. (1995). Towards a methodology for building ontologies.
[60] Unger, C., Bühmann, L., Lehmann, J., Ngonga Ngomo, A. C., Gerber, D., & Cimiano, P. (2012, April). Template-based question answering over RDF data. In Proceedings of the 21st international conference on World Wide Web (pp. 639-648). ACM.
[61] Vysniauskas, E., & Nemuraite, L. (2006). Transforming ontology representation from OWL to relational database. Information technology and control, 35(3).
[62] Wang, D., & Nyberg, E. (2015). A long short-term memory model for answer sentence selection in question answering. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 2: Short Papers) (Vol. 2, pp. 707-712).
[63] Wang, J. Y., Kuo, M. F., Han, J. C., Shih, C. C., Chen, C. H., Lee, P. C., & Tsai, R. T. H. (2017). A Telecom-Domain Online Customer Service Assistant Based on Question Answering with Word Embedding and Intent Classification. Proceedings of the IJCNLP 2017, System Demonstrations, 17-20.
[64] Weizenbaum, J. (1966). ELIZA—a computer program for the study of natural language communication between man and machine. Communications of the ACM, 9(1), 36-45.
[65] Williams, J., Raux, A., & Henderson, M. (2016). The dialog state tracking challenge series: A review. Dialogue & Discourse, 7(3), 4-33.
[66] World Trade Organization, 2018, “Trade-Related Aspects of Intellectual Property Rights,” available at: https://wto.org/trips
[67] Yih, S. W. T., Chang, M. W., He, X., & Gao, J. (2015). Semantic parsing via staged query graph generation: Question answering with knowledge base. Microsoft
[68] Yih, W. T., He, X., & Meek, C. (2014). Semantic parsing for single-relation question answering. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) (Vol. 2, pp. 643-648).
[69] Yao, X., Berant, J., & Van Durme, B. (2014). Freebase QA: Information Extraction or Semantic Parsing?. In Proceedings of the ACL 2014 Workshop on Semantic Parsing (pp. 82-86).
[70] Young, S. J. (2000). Probabilistic methods in spoken–dialogue systems. Philosophical Transactions of the Royal Society of London A: Mathematical, Physical and Engineering Sciences, 358(1769), 1389-1402.
[71] Young, S., Schatzmann, J., Weilhammer, K., & Ye, H. (2007, April). The hidden information state approach to dialog management. In Acoustics, Speech and Signal Processing, 2007. ICASSP 2007. IEEE International Conference on (Vol. 4, pp. IV-149). IEEE.
[72] Zhang, D., & Lee, W. S. (2003, July). Question classification using support vector machines. In Proceedings of the 26th annual international ACM SIGIR conference on Research and development in informaion retrieval (pp. 26-32). ACM.
[73] Zhao, T. (2016). ReinForest: Multi-Domain Dialogue Management Using Hierarchical Policies and Knowledge Ontology.
[74] Zhou, Y., Xu, B., Xu, J., Yang, L., & Li, C. (2016, October). Compositional recurrent neural networks for Chinese short text classification. In Web Intelligence (WI), 2016 IEEE/WIC/ACM International Conference on (pp. 137-144). IEEE.
[75] 經濟部智慧財產局（2017），Available: https://www.tipo.gov.tw/
[76] 商標法（民105年11月30日）
[77] 商標法逐條釋義（101年7月1日）
[78] 商標法施行細則（107年6月7日）
[79] 商標規費收費標準（101年7月1日）
[80] 侵害著作權有關財產權爭議等，106年度臺上字第2310號（民 107 年 06 月 20 日）
[81] 陳聖介、蔡欣展、王信貿、錢炳全（2007）。結合本體知識於搜尋系統方法之研究，國立臺南大學資訊工程學系。
[82] 丁俊萍（2014）。從商標法上的「混淆誤認之虞」討論商標的保護範圍–兼比較我國與美國的判決（碩士論文）。取自台灣博碩士論文系統。
[83] 李超、柴玉梅、南曉斐、高明磊（2016）。基於深度學習的問題分類方法研究。計算機科學, 43(12), 115-119。
[84] 趙晉枚、劉孔中、黃銘傑、包志成、李恬野（民94年11月30日），《商標法整體法制暨具體修法建議之研究期末報告》，經濟部智慧財產局委託開元法律專利事務所執行研究報告。