臺灣博碩士論文加值系統

由於消費旅程的破碎化導致消費者行為越來越難以捕捉，根據調查，目前多數電商平台的商品特徵標籤是以人工的方式來制定的，本研究運用人工智慧(AI)的技術自動分析商品並直接判斷商品特徵，結合網路爬蟲程式，將熱門字作為商品的外部標籤，再將消費者的特徵與之串連，如此便基於消費的歷程對後續的購買行為做出預測，使得商品推薦系統變得更加動態與精準。
本研究以玩具電商為例，目標是建立一套商品標籤自動化及購買推薦系統，首先，先將資料庫內的數據進行清洗與整理並挑選暢銷商品，接著將暢銷商品圖片予以標準化，以圖片相似度比對小於0.90作為篩選條件，找出需要由人工訂定標籤的商品圖片，最後則是用卷積神經網路(CNN)建立影像辨識模型。以新商品進行影像辨識的測試，其精準度可達82%，意即有8成商品可直接用系統貼標籤。在網路爬蟲的部分，94個商品藉由程式共取得206個熱門字可作為動態更新的標籤。
在顧客購買行為部分，將顧客曾購買過的商品予以標籤化，再將標籤的統計次數進行排序，形成顧客行為標籤，並進行分數統計，將符合外部爬蟲標籤、未購買過或是新上市的商品優先推薦給消費者。
關鍵詞：消費者旅程破碎化、人工智慧、網路爬蟲、商品推薦系統、商品標籤、影像辨識、卷積神經網路

As consumer behavior becomes increasingly difficult to capture due to the fragmentation of the consumer journey, according to the survey, most e-commerce platforms are currently developing their product feature labels manually. This study uses artificial intelligence technology to automatically analyze products and directly determine product characteristics, combining with a web crawler program that uses popular words as external tags for products, and then links consumer characteristics to them.
The objective of this study is to establish an automated product labeling and purchase recommendation system for toy e-commerce. The first step is to clean and organize the data in the database and select the best-selling products, then standardize the images of the best-selling products, and use the image similarity ratio of less than 0.90 as the filtering criterion to find the images of products that need to be manually labeled. The new product was tested for image recognition and achieved 82% accuracy, meaning 80% of the products could be directly labeled by the system. In the web crawler section, 94 products were programmed with a total of 206 popular words that could be dynamically updated as labels.
In the section of customer purchase behavior, the products that customers have purchased are labeled, and the number of times the labels are counted is sorted to form customer behavior labels, and the scores are counted to give priority to products that meet the external crawler label, have not been purchased, or are new to the market.
Keywords：Consumer Journey Fragmentation, Artificial Intelligence, Web Crawler, Commodity Recommendation System, Product Labeling, Image Recognition, Convolutional Neural Network

摘要 i
ABSTRACT ii
誌謝 iii
目錄 iv
表目錄 x
圖目錄 xi
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機 7
1.3 研究目的 9
第二章 文獻回顧 10
2.1個人化行銷相關研究 10
2.1.1 個人化行銷之定義 10
2.1.2 個人化行銷的應用與發展 13
2.1.3 個人化行銷的資料蒐集與應用 16
2.1.4 小結 18
2.2 商品影像辨識自動標籤相關研究 19
2.2.1 商品標籤 19
2.2.2 商品影像辨識 21
2.2.3 標籤自動化 25
2.2.4小結 27
2.3 網路爬蟲相關研究 28
2.4 動態商品推薦系統相關研究 30
第三章 研究方法 32
3.1 研究架構 32
3.1.1商業理解 33
3.1.2 數據理解 33
3.1.3 數據準備 34
3.1.4 建立預測模型 36
3.1.5 預測模型評估 37
3.1.6 方案產生 38
第四章 驗證結果與分析 39
4.1 數據資料處理結果 39
4.2 數據準備結果 43
4.2.1挑取暢銷商品 43
4.2.2蒐集暢銷商品圖片 43
4.2.3暢銷商品圖片標準化 44
4.2.4圖片相似度比對 45
4.2.5人工訂定標籤 47
4.2.6外部關鍵字爬蟲 53
4.3建立預測模型結果 57
4.3.1新上市圖片標準化 57
4.3.2影像辨識精準評估 59
4.4預測模型評估結果 63
4.5方案產生結果 67
4.6小結 72
第五章 結論與建議 73
5.1 研究結論 73
5.1.1研究成果 73
5.1.2學術貢獻 74
5.1.3管理意涵 75
5.2 研究建議 76
參考文獻 78
一、中文部分 78
二、英文部分 81
附錄一、暢銷商品之人工標籤訂定 90
附錄二、外部爬蟲關鍵字標籤 155

中文部分
Fimmick(2021)。Market Insights︰CRM、DMP、CDP 是什麼？怎推動行銷？。https://www.fimmick.com/2021/02/17/market-insights-what-are-crm-dmp-cdp/。搜尋日期：2021年5月13日。

Marketing Ainotam(2020)。Martech 是什麼？五分鐘了解行銷科技。https://www.ainotam.com/what-is-martech/。搜尋日期：2020年12月14日。

Ryan Lin(2018)。一股不可忽視的經濟模式：宅經濟商機(Stay-at-Home Economy)。https://www.ipsos.com/en-tw/yigubukehushidejingjimoshi-zhaijingjishangjistay-home-economy。搜尋日期：2020年12月8日。

Tiger Lee(2020)。迎戰保護隱私權的世代！MarTech + AI 實踐行銷自動化。https://www.nununi.com/growth-hacker-martech-ai-tiger/。搜尋日期：2020年12月15日。

TransBiz 跨境電子商務智庫(2020)。在你下單前就出貨！Amazon用大數據鞏固電商龍頭 https://transbiz.com.tw/amazon-big-data/。搜尋日期：2021年5月1日。

Wavent(2020)。轉型時代來臨 潮網科技助今周刊打造高效個人化消費者旅程。https://www.wavenet.com.tw/news/%E5%80%8B%E4%BA%BA%E5%8C%96%E6%B6%88%E8%B2%BB%E8%80%85%E6%97%85%E7%A8%8B/。搜尋日期：2020年12月1日。

未來流通研究所(2020)。2020【商業數據圖解】台灣實體零售業的網路銷售大追擊。https://www.mirai.com.tw/2020-taiwan-physical-retail-industry-transformation-diagram/。搜尋日期：2020年12月12日。

未來商務(2020)。MarTech 台日聯軍國家隊成形！台灣 awoo 阿物攜手日本 Repro 聯手進軍亞洲。https://fc.bnext.com.tw/awoo-repro/。搜尋日期：2020年12月15日。

成長駭客行銷誌 SEO(2020)。行銷典範轉移：消費者行為破碎化時代下AI能做什麼。https://awoo.ai/zh-hant/blog/growth-hacker-2020-ai-opening/。搜尋日期：2021年5月4日。

江啟明(1988)。模組化行銷偵測系統，國立交通大學管理科學研 究所碩士論文。

行銷資料科學(2019)。何謂個人化行銷 Precision Marketing？個人化行銷思維進化史！https://medium.com/marketingdatascience/%E4%BD%95%E8%AC%82%E7%B2%BE%E6%BA%96%E8%A1%8C%E9%8A%B7-precision-marketing-%E7%B2%BE%E6%BA%96%E8%A1%8C%E9%8A%B7%E6%80%9D%E7%B6%AD%E9%80%B2%E5%8C%96%E5%8F%B2-fae501626904。搜尋日期：2021年4月22日。

吳柏緯、梁珮綺(2020)。疫情帶動網購業績成長 上半年銷售額年增17.5%。https://money.udn.com/money/story/5612/4757127。搜尋日期：2020年12月8日。

每日頭條(2019)。自動商品標籤：電商提升「幸福指數」的新機遇。https://kknews.cc/zh-tw/tech/9pxrzp5.html。搜尋日期：2021年5月13日。

洪錦魁(2019)。Python網路爬蟲王者歸來，深智出版。搜尋日期：2021年5月13日。

看見數據(2020)。MarTech軟體工具應用與研討：當代人類的時間本來就應該更被有效利用。https://ga.awoo.com.tw/nununi-worksop-0611/。搜尋日期：2021年5月13日。

商業服務資訊網(2016)。商業1.0~4.0的商店型態與說明https://www.facebook.com/CDRICiiS/posts/999831146771796:0。搜尋日期：2021年5月1日。

黃穎竹(2019).利用深度卷積網路訓練可應用於無人商店的多物件辨識模型 (Doctoral dissertation, fu ren da xue tong ji zi xun xue xi ying yong tong ji shuo shi ban). 搜尋日期：2021年5月13日。

酷比樂官方網站(2021)。https://www.amuzinc.com/。搜尋日期：2021年3月1日。

藍弋丰(2020)。宅經濟下誰是受益者？外送速食、零嘴、掃地機器人、芭比娃娃。https://technews.tw/2020/08/03/who-are-the-beneficiaries-under-the-housing-economy/。搜尋日期：2020年12月8日。

羅文坤(1986)。行銷傳播學，三民書局。搜尋日期：2021年5月4日。

 

英文部分
Ahmed, S. A., & d'Astous, A. (2004). Perceptions of countries as producers of consumer goods: AT‐shirt study in China. Journal of Fashion Marketing and Management: An International Journal.

Arratia, A., & Sepúlveda, E. (2019, September). Convolutional neural networks, image recognition and financial time series forecasting. In Workshop on Mining Data for Financial Applications ,60-69.

Bangwayo-Skeete PF, Skeete RW (2015) Can Google data improve the forecasting performance of tourist arrivals? A mixed-data sampling approach. Tour Manag 46:454–464Return to ref 2015 in article

Breese, J.S., Heckerman, D., & Kadie, C. (1998). Empirical analysis of predictive algorithms for collaborative filtering. Proceedings of the 14th conference on uncertainty in artificial intelligence (UAI-98) (pp. 43–52).

Brentan, B. M., Campbell, E., Meirelles, G. L., Luvizotto, E., & Izquierdo, J. (2017). Social network community detection for DMA creation: criteria analysis through multilevel optimization. Mathematical Problems in Engineering, 2017.

Chang, M. K., Cheung, W., & Tang, M. (2013). Building trust online: Interactions among trust building mechanisms. Information & management, 50(7), 439-445.

Chen, F., Deng, P., Wan, J., Zhang, D., Vasilakos, A. V., & Rong, X. (2015). Data mining for the internet of things: literature review and challenges. International Journal of Distributed Sensor Networks, 11(8), 431047.

Chen, L. J., Zhang, Z. K., Liu, J. H., Gao, J., & Zhou, T. (2017). A vertex similarity index for better personalized recommendation. Physica a: statistical Mechanics and its Applications, 466, 607-615.

Chen, W., Ning, H., Mao, X., & Wang, B. (2008, October). Data mining modeling for electromagnetic scattering computing. In 2008 Fifth International Conference on Fuzzy Systems and Knowledge Discovery (Vol. 2, pp. 137-140). IEEE.

Chen, X. W., & Lin, X. (2014). Big data deep learning: challenges and perspectives. IEEE access, 2, 514-525.

Chen, X., Au, W. M., & Li, K. (2004). Consumption of children's wear in a big city in Central China: Zhengzhou. Journal of Fashion Marketing and Management: An International Journal.

Costa-Montenegro, E., Barragáns-Martínez, A. B., & Rey-López, M. (2012). Which App? A recommender system of applications in markets: Implementation of the service for monitoring users’ interaction. Expert systems with applications, 39(10), 9367-9375.

Digital Commerce 360 estimates. (2021). US ecommerce grows 44.0% in 2020. https://www.digitalcommerce360.com/article/us-ecommerce-sales/

Du, W., Ning, H., Wei, Y., & Wang, J. (2013, August). Fuzzy double-threshold track association algorithm using adaptive threshold in distributed multisensor-multitarget tracking systems. In 2013 IEEE International Conference on Green Computing and Communications and IEEE Internet of Things and IEEE Cyber, Physical and Social Computing (pp. 1133-1137). IEEE.

Eckerson, W., & Watson, H. (2000). Harnessing customer information for strategic advantage: technical challenges and business solutions. Industry study, 6.

El Khaddar, M. A., & Boulmalf, M. (2017). Smartphone: the ultimate IoT and IoE device. Smartphones from an applied research perspective, 137.

Fickel, L. (1999). Know Your Customer: Intimacy, relationships, one-to-one-it's a giant lovefest out there. CIO-FRAMINGHAM MA-, 12, 62-75.

Fiedler, N., Bestmann, M., & Hendrich, N. (2018, June). Imagetagger: An open source online platform for collaborative image labeling. In Robot World Cup (pp. 162-169). Springer, Cham.

Gartner. (2019). Gartner Says Marketers Utilize Only 58% of Their Martech Stack’s Potential. https://www.gartner.com/en/newsroom/press-releases/2019-11-18-gartner-says-marketers-utilize-only-58--of-their-mart

Gartner. (2021). Customer Relationship Management (CRM). https://www.gartner.com/en/information-technology/glossary/customer-relationship-management-crm

Geraldine G. Fennell (1978, April). Perceptions of the product-in-use situation (pp. 39-47)

Girshick, R. (2015). Fast r-cnn. In Proceedings of the IEEE international conference on computer vision (pp. 1440-1448).

Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2014). Rich feature hierarchies for accurate object detection and semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 580-587).

Greenberg, P. (2001). CRM at the speed of light: Capturing and keeping customers in Internet real time. McGraw-Hill Professional.

Gu, J., Wang, Z., Kuen, J., Ma, L., Shahroudy, A., Shuai, B., Liu, T., Wang, X., Wang, G., Cai, J., Chen, T. (2018). Recent advances in convolutional neural networks. Pattern Recognition, 77, 354-377.Guo, B., Yu, Z., Zhang, D., & Zhou, X. (2014). Cross-community sensing and mining. IEEE Communications Magazine, 52(8), 144-152.

Hao, F., Li, S., Min, G., Kim, H. C., Yau, S. S., & Yang, L. T. (2015). An efficient approach to generating location-sensitive recommendations in ad-hoc social network environments. IEEE Transactions on Services Computing, 8(3), 520-533.

Hao, F., Yau, S. S., Min, G., & Yang, L. T. (2014). Detecting k-balanced trusted cliques in signed social networks. IEEE internet computing, 18(2), 24-31.

Hosseinmardi, H., Mattson, S. A., Rafiq, R. I., Han, R., Lv, Q., & Mishra, S. (2015, December). Analyzing labeled cyberbullying incidents on the instagram social network. In International conference on social informatics (pp. 49-66). Springer, Cham.

Hsu, H. J., & Burns, L. D. (2002). Clothing evaluative criteria: A cross-national comparison of Taiwanese and United States consumers. Clothing and Textiles Research Journal, 20(4), 246-252.

Hu, L., Lin, K., Hassan, M. M., Alamri, A., & Alelaiwi, A. (2015). CFSF: On cloud-based recommendation for large-scale E-commerce. Mobile Networks and Applications, 20(3), 380-390.

Iglesias, O., Singh, J. J., Casabayó, M., Hung, K. P., Chen, A. H., Peng, N., ... & Chou, C. L. (2011). Antecedents of luxury brand purchase intention. Journal of Product & Brand Management.

Joung, J., & Kim, H. M. (2020). An LDA-based Approach for Product Attribute Identification from Online Customer Reviews.

Ke, G., Du, H. L., & Chen, Y. C. (2021). Cross-platform dynamic goods recommendation system based on reinforcement learning and social networks. Applied Soft Computing, 104, 107213.

Khodabandehlou, S., & Rahman, M. Z. (2017). Comparison of supervised machine learning techniques for customer churn prediction based on analysis of customer behavior. Journal of Systems and Information Technology.

Koohi, H., & Kiani, K. (2017). A new method to find neighbor users that improves the performance of collaborative filtering. Expert Systems with Applications, 83, 30-39.

Lai, X. A. (2009, April). Segmentation study on enterprise customers based on data mining technology. In 2009 First International Workshop on Database Technology and Applications (pp. 247-250). IEEE.

LeCun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11), 2278-2324.

Lee, C. Y., & Ke, H. R. (2012). A Study on User Perceptions and User Behavior of an Online Federated Search System. Journal of Educational Media & Library Sciences, 49(3).

Li, H., Lin, K., & Li, K. (2011). Energy-efficient and high-accuracy secure data aggregation in wireless sensor networks. Computer Communications, 34(4), 591-597.

Lin, S. W., & Lo, L. Y. S. (2015). Mechanisms to motivate knowledge sharing: integrating the reward systems and social network perspectives. Journal of Knowledge Management.

LOTAME,2019. What is a Data Management Platform (DMP)? https://www.lotame.com/what-is-a-data-management-platform/

Miklosik, A., Kuchta, M., & Zak, S. (2018). Monetising Content through Delivery of Advertisements: The Case of Ad Blockers. Ad Alta: Journal of Interdisciplinary Research, 8(1).

Miklosik, A., Kuchta, M., Evans, N., & Zak, S. (2019). Towards the adoption of machine learning-based analytical tools in digital marketing. IEEE Access, 7, 85705-85718.

Mohammadi, V., Rahmani, A. M., Darwesh, A. M., & Sahafi, A. (2019). Trust-based recommendation systems in Internet of Things: a systematic literature review. Human-centric Computing and Information Sciences, 9(1), 1-61.

Mulvenna, M. D., Anand, S. S., & Büchner, A. G. (2000). Personalization on the net using web mining: introduction. Communications of the ACM, 43(8), 122-125.

Nicosia, F. M., & Mayer, R. N. (1976). Toward a sociology of consumption. Journal of consumer research, 3(2), 65-75.

Oltman, J., & Zabin, J. (2002). Precision Marketing: More Science Than Fiction. Seurat Company. CEO”, Sterling Publications, London.

Park, D. H., Kim, H. K., Choi, I. Y., & Kim, J. K. (2012). A literature review and classification of recommender systems research. Expert systems with applications, 39(11), 10059-10072.

Peng, D., Li, T., Wang, Y., & Chen, C. P. (2018, June). Research on information collection method of shipping job hunting based on web crawler. In 2018 Eighth international conference on information science and technology (ICIST) (pp. 57-62). IEEE.

Rahman, O., Zhu, X., & Liu, W. S. (2008). A study of the pyjamas purchasing behaviour of Chinese consumers in Hangzhou, China. Journal of Fashion Marketing and Management: An International Journal.

Ren, S., He, K., Girshick, R., & Sun, J. (2015). Faster r-cnn: Towards real-time object detection with region proposal networks. In Advances in neural information processing systems(pp. 91-99).

Rose, N. (1988). Calculable minds and manageable individuals. History of the Human Sciences 1 (2): 179-200.

Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., ... & Fei-Fei, L. (2015). Imagenet large scale visual recognition challenge. International journal of computer vision, 115(3), 211-252.

S. Xin, D. Hua, K. Xiaoyun and Z. Mingming. (2017). Scrapy-based Industrial Vulnerability Reptile Design, Cyberspace Security, 66-71.

Schoenbachler, D. D., & Gordon, G. L. (2002). Trust and customer willingness to provide information in database-driven relationship marketing. Journal of interactive marketing, 16(3), 2-16.

Shukla, P., & Purani, K. (2012). Comparing the importance of luxury value perceptions in cross-national contexts. Journal of Business Research, 65(10), 1417-1424.

Sirgy, M. J. (1982). Self-concept in consumer behavior: A critical review. Journal of consumer research, 9(3), 287-300.

Solomon, M. R. (1983). The role of products as social stimuli: A symbolic interactionism perspective. Journal of Consumer research, 10(3), 319-329.

Tettamanzi, A. G., Carlesi, M., Pannese, L., & Santalmasi, M. (2007, April). Business intelligence for strategic marketing: Predictive modelling of customer behaviour using fuzzy logic and evolutionary algorithms. In Workshops on Applications of Evolutionary Computation (pp. 233-240). Springer, Berlin, Heidelberg.

Tjiptono, Fandy. (2005). Manajemen Jasa, Edisi kedua. Yogyakarta: Andi

Wallendorf, M., & Reilly, M. D. (1983). Ethnic migration, assimilation, and consumption. Journal of Consumer Research, 10(3), 292-302.

Wedel, M., & Kannan, P. K. (2016). Marketing analytics for data-rich environments. Journal of Marketing, 80(6), 97-121.

Wu, J., & Delong, M. (2006). Chinese perceptions of western‐branded denim jeans: a Shanghai case study. Journal of Fashion Marketing and Management: An International Journal.

Wyner, G. A. (2001). When 360 degrees is not enough. Marketing management, 10(2), 4.

Xiao, J., Wang, M., Jiang, B., & Li, J. (2018). A personalized recommendation system with combinational algorithm for online learning. Journal of Ambient Intelligence and Humanized Computing, 9(3), 667-677.

Yang, Y., Guo, B., Yu, Z., & He, H. (2013, December). Social activity recognition and recommendation based on mobile sound sensing. In 2013 IEEE 10th International Conference on Ubiquitous Intelligence and Computing and 2013 IEEE 10th International Conference on Autonomic and Trusted Computing (pp. 103-110). IEEE.

You, Z., Si, Y. W., Zhang, D., Zeng, X., Leung, S. C., & Li, T. (2015). A decision-making framework for precision marketing. Expert Systems with Applications, 42(7), 3357-3367.

Yu, C., Yu, T. F., & Cheng, Y. S. (2012). The investigation under technology acceptance model for the factors of using e-book among employees of a Taiwanese financial and insurance company. International Journal of Digital Content Technology and its Applications, 6(3).

Yu, P. S. (1999, April). Data mining and personalization technologies. In Proceedings. 6th International Conference on Advanced Systems for Advanced Applications (pp. 6-13). IEEE.

Z. Lizhu and L. Ling. (2005). Research on Focusing Crawler Technology, Computer application, 1965-1969.