臺灣博碩士論文加值系統

　　客服在顧客關係管理扮演著與顧客互動的功能及企業與顧客之間重要的橋樑角色。當今客服須有一定的專業技術與經驗，方能協助顧客解決問題，故需花費時間進行教育訓練與累積經驗，而處理的問題大多為重複性問題；且客服是企業與顧客接觸的第一線人員，若客服人員因經驗不足無法處理問題，將直接影響到企業形象與價值。本研究針對資訊服務業客服，建置基於FAQ之問答模型，並為了使客戶直覺的反應與解決問題，並以線上客服機器人之方式實現其模型，讓使用者能以自然語言的方式敘述其問題，且與機器人對話的方式亦增加了互動性。為了能讓系統更快速且準確找尋答案，先針對使用者問句做問題分類，再於該分類下找答案，問題分類以最大熵學習法訓練模型，找尋答案以餘弦相似度計算使用者問句與FAQ問句，相似度最高即為最佳解答。本研究問答資料來源為E公司，問答系統最後將由公司的專業人員實測，並探討其效益。

　Customer Service plays an important role in the customer relationship management (CRM) as a bridge between enterprises and clients .Nowadays,customer service must have certain professional technology and experience, in order to help customers solve the problem, they need to spend time to education training and accumulated experience,and most of the problems and handle the repeatability problem.Customer service is the first line personnel that enterprises contact with customers.If customer service personnel are unable to deal with problems due to lack of experience, it will directly affect the image and value of enterprises.This study aims at customer service in information service industry, develops a question-and-answer model based on FAQ, and in order to make customers'intuitive response and problem-solving,and implement the model in the way of customer service robots, so that users can narrate their problems in natural language, and the way of dialogue with robots also increases the interaction.In order to make the system search for answers more quickly and accurately, firstly, classify the questions according to the users'questions, and train the model with the Maximum Entropy Algorithm.Using cosine similarity to calculate user questions and FAQ questions, the best answer is the highest similarity.The data source of this study is Company E. The QA system will be tested by the company's professionals and its benefits will be discussed.

摘　要………….....................................................................................i
ABSTRACT……………………………………………………………………………………….ii
誌　謝………….....................................................................................iv
目　錄………….....................................................................................v
表目錄………….....................................................................................viii
圖目錄………….....................................................................................ix
第一章 緒論………….........................................................................1
1.1 研究背景與動機…………..............................................................1
1.2 研究目的…………..........................................................................2
1.3 研究流程…………..........................................................................2
第二章 文獻探討………….................................................................5
2.1 客戶支援系統…………..................................................................5
2.1.1 問題分類之客服系統…………...................................................5
2.1.2 FAQ系統…………........................................................................7
2.1.3 問答系統………….......................................................................8
2.2 演算法…………..............................................................................10
2.2.1 支援向量機（Support Vector Machine, SVM）……………………10
2.2.2 最大熵學習法（Maximum Entropy, ME）…………………………..12
2.3 文字探勘.....................................................................................13
2.3.1 文字探勘前處理......................................................................13
2.3.2文字探勘之應用.......................................................................15
第三章 研究方法………………….........................................................18
3.1 問題定義與假設.........................................................................18
3.2 研究架構.....................................................................................21
3.3 問答模型建置.............................................................................23
3.3.1 詞庫建置..................................................................................23
3.3.2 擷取特徵值..............................................................................24
3.3.3 問答模型訓練..........................................................................24
3.4 最大熵演算法求解步驟.............................................................24
3.5 語意框架.....................................................................................27
3.6 相似度計算.................................................................................28
第四章 系統建構與驗證................................................................30
4.1系統建構環境與工具..................................................................30
4.1.1系統開發環境...........................................................................30
4.1.2系統技術架構...........................................................................31
4.2系統建置......................................................................................32
4.2.1詞庫模組...................................................................................33
4.2.2分類器訓練...............................................................................34
4.2.3知識庫建置................................................................................36
4.2.4特徵關鍵字篩選與相似度計算................................................37
4.3系統介面與操作說明...................................................................39
4.4系統驗證.......................................................................................44
4.4.1分類矩陣....................................................................................44
4.4.2問題分類實驗結果及效能評估................................................45
4.5.3問題答案實驗結果....................................................................47
第五章 結論與建議.........................................................................50
5.1結論及研究成果...........................................................................50
5.2未來發展建議...............................................................................51
參考文獻.............................................................................................52
附錄A………….......................................................................................58

Chen, G., & Ge, Z. (2019, June). SVM-tree and SVM-forest algorithms for imbalanced fault classification in industrial processes. IFAC Journal of Systems and Control, 8. doi:https://doi.org/10.1016/j.ifacsc.2019.100052
Chen, K.-J., & Liu, S.-H. (1992). Word identification for Mandarin Chinese sentences. Paper presented at the Proceedings of the 14th conference on Computational linguistics-Volume 1.
Darroch, J. N., & Ratcliff, D. (1972). Generalized iterative scaling for log-linear models. The annals of mathematical statistics, 43(5), 1470-1480.
Dawson, K. (2001). The Call Center Handbook1: The Complete Guide to Starting, Running, and Improving Your Call Center: CRC Press.
Della Pietra, S., Della Pietra, V., Mercer, R. L., & Roukos, S. (1992). Adaptive language modeling using minimum discriminant estimation. Paper presented at the Acoustics, Speech, and Signal Processing, 1992. ICASSP-92., 1992 IEEE International Conference on.
Demeter, K., Szász, L., & Kő, A. (2019, March). A text mining based overview of inventory research in the ISIR special issues 1994–2016. International Journal of Production Economics, 209, 134-146. doi:https://doi.org/10.1016/j.ijpe.2018.06.006
Dumais, S., Platt, J., Heckerman, D., & Sahami, M. (1998). Inductive learning algorithms and representations for text categorization. Paper presented at the Proceedings of the seventh international conference on Information and knowledge management.
Fan, C.-K., & Tsai, W.-H. (1988). Automatic word identification in Chinese sentences by the relaxation technique. Computer Processing of Chinese Oriental Languages, 4(1), 35-56.
Galati, F., & Bigliardi, B. (2019, August). Industry 4.0: Emerging themes and future research avenues using a text mining approach. Computers in Industry, 109, 100-113. doi:https://doi.org/10.1016/j.compind.2019.04.018
Gandomi, A. H., Yang, X. S., Talatahari, S., & Alavi, A. H. (2013). Metaheuristic Algorithms in Modeling and Optimization. Metaheuristic Applications in Structures and Infrastructures(pp. 1-24): Elsevier.
Gzyl, H., Ter Horst, E., & Molina, G. (2015, November). Application of the method of maximum entropy in the mean to classification problems.Application of the method of maximum entropy in the mean to classification problems, 437, 101-108. doi:https://doi.org/10.1016/j.physa.2015.05.105
Kao, A., & Poteet, S. (2005). Text mining and natural language processing: introduction for the special issue. ACM SIGKDD Explorations Newsletter, 7(1), 1-2.
Liu, J. W. (2019, February). Using big data database to construct new GFuzzy text mining and decision algorithm for targeting and classifying customers. Computers & Industrial Engineering, 128, 1088-1095. doi:https://doi.org/10.1016/j.cie.2018.04.003
Lu, W.-H., Chien, L.-F., & Lee, H.-J. (2002). Translation of web queries using anchor text mining. J ACM Transactions on Asian Language Information Processing, 1(2), 159-172.
Moreo, A., Eisman, E. M., Castro, J. L., & Zurita, J. M. (2013, November). Learning regular expressions to template-based FAQ retrieval systems. Knowledge-Based Systems, 53, 108-128. doi:https://doi.org/10.1016/j.knosys.2013.08.018
Moreo, A., Navarro, M., Castro, J. L., & Zurita, J. M. (2012). A high-performance FAQ retrieval method using minimal differentiator expressions. Knowledge-Based Systems, 36, 9-20. doi:ttps://doi.org/10.1016/j.knosys.2012.05.015
Nie, J.-Y., Hannan, M.-L., & Jin, W. (1995). Unknown word detection and segmentation of Chinese using statistical and heuristic knowledge. Communications of COLIPS, 5(1), 47-57.
Peral, J., Ferrández, A., Gregorio, E. D., Trujillo, J., Maté, A., & Ferrández, L. J. (2015, March). Enrichment of the phenotypic and genotypic Data Warehouse analysis using Question Answering systems to facilitate the decision making process in cereal breeding programs. Ecological Informatics, 26, 203-216.
doi:https://doi.org/10.1016/j.ecoinf.2014.05.003
Rao, Y., Xie, H., Li, J., Jin, F., Wang, F. L., & Li, Q. (2016, December). Social emotion classification of short text via topic-level maximum entropy model. Information & Management, 53(8), 978-986. doi:https://doi.org/10.1016/j.im.2016.04.005
Salton, G., & Buckley, C. (1988). Term-weighting approaches in automatic text retrieval. Information processingmanagement, 24(5), 513-523.
Salton, G., & McGill, M. J. (1986). Introduction to modern information retrieval.
Sebastiani, F. (2005). Text Categorization Text Mining and Its Applications, 109-129.
Sparck Jones, K. (1972). A statistical interpretation of term specificity and its application in retrieval. Journal of documentation, 28(1), 11-21.
Spengler, B. (1999). 1999年度台灣業者之顧客關係管理應用現狀調查. 電子化企業經理人報告(3), 9-15.
Sullivan, D. (2001). Document warehousing and text mining: techniques for improving business operations, marketing, and sales: John Wiley & Sons, Inc.
Sun, G., Liang, L., Chen, T., Xiao, F., & Lang, F. (2018, July). Network traffic classification based on transfer learning. Computers & Electrical Engineering, 69, 920-927. doi:https://doi.org/10.1016/j.compeleceng.2018.03.005
Takeuchi, H., Masuda, S., Miyamoto, K., & Akihara, S. (2018). Obtaining Exhaustive Answer Set for Q&A-based Inquiry System using Customer Behavior and Service Function Modeling. Procedia Computer Science, 126, 986-995. doi:https://doi.org/10.1016/j.procs.2018.08.033
Wang, L. L., Ngan, H. Y. T., & Yung, N. H. C. (2018, October). Automatic incident classification for large-scale traffic data by adaptive boosting SVM. Information Sciences, 467, 59-73. doi:https://doi.org/10.1016/j.ins.2018.07.044
Yang, X. S. (2014). Nature-Inspired Optimization Algorithms. London: Elsevier, 15.
Yang, Y., Xu, D. L., Yang, J. B., & Chen, Y. W. (2018, December). An evidential reasoning-based decision support system for handling customer complaints in mobile telecommunications. Knowledge-Based Systems, 162, 202-210.
doi:https://doi.org/10.1016/j.knosys.2018.09.029
Yang, Z., & Su, X. (2012). Customer Behavior Clustering Using SVM. Physics Procedia, 33, 1489-1496. doi:https://doi.org/10.1016/j.phpro.2012.05.243
Vapnik, V. (1995). The nature of statistical learning theory. New York: John Wiley & Sons.
Zhang, W., Yoshida, T., & Tang, X. (2011). A comparative study of TF* IDF, LSI and multi-words for text classification. Expert Systems with Applications, 38(3), 2758-2765.
Zhao, L. T., & Zeng, G. R. (2019, February). Analysis of Timeliness of Oil Price News Information Based on SVM. Energy Procedia, 158, 4123-4128.
doi:https://doi.org/10.1016/j.egypro.2019.01.821
尹其言、楊建民（2010）。應用文件分群與文字探勘技術於機器學習領域趨勢分析以 SSCI 資料庫為例。長榮大學學報， 14（2），頁 1-16。 doi: 10.30115
王玮、严文涛、苏琦、刘荫、于展鹏、殷齐林（2017）。客服运维智能人机对话系统设计。烟台职业学院学报，2，頁 83-85。
吳佩紋（2011）。FAQ自動分類系統與使用者滿意度之研究（未出版之碩士論文）。淡江大學資訊管理學系碩士班，新北市。
呂宛靜（2013）。應用探勘技術於內容管理系統之研究-以電子期刊審稿系統開發為例。正修科技大學資訊管理研究所碩士論文，高雄市。 取自https://hdl.handle.net/11296/kxnh97
周雅倩（2004）。最大熵方法及其在自然语言处理中的应用（博士論文）。取自http://cdmd.cnki.com.cn/Article/CDMD-10246-2005121138.htm
林盈宏、陳毅晟、莊凱翔、許乙清（2017）。結合 Jieba 與 k-NN 於政府開放資料集推薦之臉書貼文。TANET2017 臺灣網際網路研討會，頁 938-943。
施鳳美（2006）。應用案例式推理探討軟體技術問題診斷之研究（未出版之碩士論文）。輔仁大學資訊管理學系，新北市。
張凱鈞（2015）。問答系統中答案型態辨識之研究（未出版之碩士論文）。元智大學資訊管理學系，桃園縣。
張源福（2008）。網路大學自動回覆客服系統之規劃（未出版之碩士論文）。崑山科技大學資訊管理研究所，台南市。
郭佳鑫（2011）。線上學習資源自動分類系統之研究（未出版之碩士論文）。世新大學資訊管理學研究所(含碩專班)，臺北市。
陳如妙（2003）。應用文件探勘技術於FAQ系統之建置（未出版之碩士論文）。銘傳大學資訊管理研究所，台北市。
陳冠憲（2009）。利用文件探勘技術來自動分類民眾陳情文件（未出版之碩士論文）。國立中山大學資訊管理學系研究所，高雄市。
陳盈如（2004）。稅務法規問答系統之研究（未出版之碩士論文）。銘傳大學資訊管理學系碩士班，台北市。
劉正廉（2014）。運用案例式推理機制建構電腦問題診斷專家系統 -以個案公司為例（未出版之碩士論文）。國立交通大學管理學院資訊管理學程，新竹市。
劉康宇（2010）。客訴服務知識管理系統於電子科技業之應用（未出版之碩士論文）。國立臺北科技大學工業工程與管理研究所，台北市。
鄭雅惠（2012）。客服中心顧客問題分類方法之研究（未出版之碩士論文）。世新大學資訊管理學研究所(含碩專班)，臺北市。
蕭惠如（2017）。應用文字探勘於資訊管理領域研究趨勢（未出版之碩士論文）。銘傳大學資訊管理學系碩士在職專班，台北市。