臺灣博碩士論文加值系統

隨著網際網路經濟的發展，各家在線旅遊公司網站上，累積了眾多消費者對於各式旅館體驗後的評論，使用者不再只是被動的接收訊息，而是主動的成為訊息的傳播者。每當旅客完成一趟旅程體驗，都可產生出大量的數據資料，舉凡：抵達時間、餐點飲食、客戶服務等，這些資料皆可提供予旅館經營者，透過大數據分析策略，藉此提升旅館的服務品質。然而，大數據分析所帶來的挑戰，不僅侷限於數據的數量與分析的速度，連同數據的來源與其型態、種類，都得考量進分析過程中。本研究將建立基於本體論為核心的大數據語意分析雲端平台，藉由顧客住宿體驗後的反饋評論，建構以客戶滿意度為衡量基礎的指標評量系統。此平台不僅能讓旅館從業人員有效率地擷取與儲存網路評論訊息，還能透過以國際標準為知識核心的本體論推論引擎，找出旅館之特點與定位方向。

Hospitality industry is a data rich industry that captures huge volumes of data of various types including arrival time, the frequency of use of public facilities, meals diet, customer service, social network comments, etc. with high velocity. These data encapsulate useful information regarding every phase of the customer journey. The effective use of analytics can improve dramatically how business is run in terms of delivering memorable and personalized guest experiences, while maximizing revenue and profits. The challenge of Big Data doesn't just stem from the volume and velocity of the data sets themselves, but also from the variety challenge posed by gaining big data insight in the context of an industry. For example, how to score the performance of hotels based on a variety of semantic data sources. Ontology is a formal representation of knowledge as a hierarchy of concepts within a domain, using a shared vocabulary to denote the types, properties and interrelationships of those concepts. An ontology driven Big Data analytics has potential in providing a pragmatic framework to address the semantic challenges presented by Big Data sets. This research aims to develop an ontology driven Big Data Sematic Analytic Platform which enables hotels to improve their overall performance in furnishing better customer experiences by infusing analytics through every phase of the guest journey. The platform allows the business in hospitality industry not only to capture and store the influx of semantic data effectively but also to evaluate its key performance factors by using ontology organized international standard knowledge-based such as U.S. AAA Diamond evaluation systems.

Contents
Chapter 1 Introduction 1
1.1 Background and motivation 1
1.2 Research objectives 4
1.3 Research Overview 5
Chapter 2 Literature Comment 6
2.1 Big data 6
2.2 The American Diamond Evaluation System 6
2.3 Chinese Word Segmentation 7
2.4 Web 2.0 8
2.5 Fake Comments 8
2.6 Ontology 10
Chapter 3 Research Methodology 11
3.1 Research architecture 11
3.2 Data capture 12
3.2.1 Establish the target hotel 13
3.2.2 Customers comments 14
3.2.3 Customers comments clean up 14
3.2.4 Filter fake comments 16
3.3 Data processing 17
3.3.1 Lexicon collection 17
3.3.2 Word processing, stop word processing 18
3.3.3 Characteristic word analysis 19
3.3.4 Build evaluation ontology engine 20
3.3.5 Build customer comments semantic analysis, evaluation platform 21
Chapter 4 Research result 22
4.1 Data Capture 22
4.2 Language Detection 23
4.2.1 Detect non-Chinese comments 23
4.2.2 Detection of non-traditional Chinese comments 24
4.3 Data cleansing 25
4.4 Fake Comments Detection 26
4.4.1 Delete the comments that are from the same person. 26
4.4.2 Delete the comments that has shown first-person more than 6 times 27
4.5 Word Segmentation 28
4.6 Classification Model 28
4.7 Comment Analysis Engine 29
4.8 User Interface 33
Chapter 5 Conclusion 35
References 36
Appendix I 40
1. Halfwidth to fullwidth 40
2. The Remove Symbol 40
3. The Remain Symbol 42
4. Emoticons Meaning 42
Appendix II 44
1. The Dictionary of Jieba Participle 44
Appendix III 45
1. Exterior 45
2. Public Area 45
3. Other Area 45
4. Room 46
5. Bathroom Facilities 46

[1]雪球-美國旅遊巨頭是怎麼賺錢的?( https://xueqiu.com/9041141730/39046954)
[2] [SAS] 趨勢觀察／善用大數據 決策更精準(http://www.sasresource.com/artical374.html)
[3] Abhishek Singh. Big Data, Multimedia, Sentiment Analysis & Monetization: http://www. infosysblogs.com/bigdata/2013/02/big_data_multimedia.html, Access date: 2016/09/29.
[4] Mark van Rijmenam. Why Hotels Should Apply Big Data Analytics To Provide a Unique Guest Experience: https://datafloq.com/read/hotels-should-apply-big-data- analytics-guest/81, Access date: 2016/09/29.
[5]Big Security for Big Data: Addressing Security Challenges for the Big Data Infrastructure- Yuri Demchenko,Canh Ngo,Cees de Laat,Peter Membrey,Daniil Gordijenko
[6] Xiang Z., Schwartz Z., Gerdes J.H., & Uysal M. What can big data and text analytics tell us about hotel guest experience and satisfaction? International Journal of Hospitality Management, 44, 120-130, 2015.
[7] Bjørkelund E., Burnett T.H., & Nørvåg K. A study of opinion mining and visualization of hotel comments. Paper presented at the ACM International Conference Proceeding Series, 229-238, 2012.
[8] Zhao Y., Dong S., & Yang J. Effect research of aspects extraction for chinese hotel comments based on machine learning method. International Journal of Smart Home, 9(3), 23-34, 2015.
[9] Barbosa R.R.L., Sánchez-Alonso S., & Sicilia-Urban M.A. Evaluating hotels rating prediction based on sentiment analysis services. Aslib Journal of Information Management, 67(4), 392-407, 2015.
[10] Liu B., Hu M., and Cheng J. Opinion observer: analyzing and comparing opinions on the Web. In Proceedings of the 14th international conference on World Wide Web (WWW '05), 342-351, 2005.
[11] 網路旅行社OTA有哪些優勢？從實體到虛擬的電子化進程(https://www.stockfeel.com.tw/%E7%B6%B2%E8%B7%AF%E6%97%85%E8%A1%8C%E7%A4%BEota%E6%9C%89%E5%93%AA%E4%BA%9B%E5%84%AA%E5%8B%A2%EF%BC%9F%E5%BE%9E%E5%AF%A6%E9%AB%94%E5%88%B0%E8%99%9B%E6%93%AC%E7%9A%84%E9%9B%BB%E5%AD%90%E5%8C%96/)
[12] Hashem, I. A. T., Yaqoob, I., Anuar, N. B., Mokhtar, S., Gani, A., & Ullah Khan, S. (2015).The rise of “big data” on cloud computing: Comment and open research issues. InformationSystems,47, 98-115. doi:10.1016/j.is.2014.07.006
[13]中文互聯網數據資訊中心-旅遊O2O發展的基礎之一:PMS現狀及發展趨勢(http://www.199it.com/archives/407470.html)
[14]數位時代-一次搞懂大數據, (https://www.bnext.com.tw/article/35807/bn-2015-03-31-151014-36, 2015.04.01)

[15] 林大貴(2016)，Hadoop+Spark大數據分析與機器學習整合開發實戰(初版五 刷)，博碩文化股份有限公司
[16]AAA newsroom, American Automobile Association
http://newsroom.aaa.com/diamond-ratings/diamond-rating-definitions/
[17]美國鑽石AAA
http://elitegen.singtao.ca/%E6%97%85%E9%81%8A/%E9%85%92%E5%BA%97%E4%B8%96%E7%95%8C%E7%9A%84%E9%91%BD%E7%9F%B3/
[18] 結巴(2012/10/07). https://github.com/fxsjy/jieba. Access date: 2016/12/17.
[19] Tim O'Reilly(2005/09/30). What Is Web 2.0-Design Patterns and Business Models for the Next Generation of Software. http://www.oreilly.com/pub/a/web2/archive/ what-is-web-20.html. Access date: 2016/09/30.
[20]Yogev Vaknin & Nir Hemed, March 2, 2013 - Identifying Deceptive Hotel Reviews
[21] Feng S., Banerjee R. and Choi Y. Syntactic stylometry for deception detection. 50th Annual Meeting of the Association for Computational Linguistics, ACL 2012 - Proceedings of the Conference, 2, pp. 171-175. 2012.
[22] Mukherjee A., Venkataraman V., Liu B. and Glance N. What yelp fake comment filter might be doing? Proceedings of the 7th International Conference on Weblogs and Social Media, ICWSM 2013, pp. 409-418. 2013.
[23] Kyung-Hyan Yoo & Ulrike Gretzel. Comparison of Deceptive and Truthful Travel Comments,2009
[24]Tsianos,N.,Germanalos,P.,Belk,M.,Lekkas,Z.,Samaras,G.,andMourlas,C.(2013),An individual differences approach in designing ontologies for efficient personalization, Vol.418.
[25]Horridge,M., and Brandt, S.(2011)A Practical Guide To Building OWL Ontologies Using Prot’eg’e 4 and CO-ODE Tools Edition 1.3. The University Of Manchester
[26] eBiz, Top 15 Most Popular Travel Websites, http://www.ebizmba.com/articles/
travel-websites, Access date: 2017/05
[27]中文百科在線-停止詞(http://www.zwbk.org/zh-tw/Lemma_Show/257895.aspx)
[28] Chi, Y. L., A consumer-centric design approach to develop comprehensive Knowledge-based systems for keyword discovery, Expert Systems with Applications, 2008.
[29] Ku, L., W. & Chen, H., H, Mining opinions from the web: beyond relevance retrieval. Journal of American Society for Information Science and Technology, 58(12), 1838-850. 2007.
[30] 中央研究院，“中文斷詞系統”，http://ckipsvr.iis.sinica.edu.tw/ ，Access date: 2016/09/29.
[31]張甜，“辨識虛假評論之研究 – 以旅館業為例”，2014.
[32] 洪旭典, “論批踢踢實業坊中的表情符號使用” , 2013