以往的研究主要在同個領域裡做推薦系統，少數跨領域之研究運用轉移學習方法做跨領域推薦，大部分也都在性質類似的領域間建立網路，因此本研究將嘗試設計一個針對社群媒體的跨領域推薦系統，對Instagram的使用者圖像貼文紀錄進行圖像分析，並與其他領域建立異質性網路，並推薦給Instagram使用者另一個領域內可能有興趣的資訊。
本研究以Instagram為目標領域，爛番茄影評網站為來源領域，採用餘弦相似度推薦方法與類協同過濾推薦方法，我們透過Instagram使用者和電影評論者之間的餘弦相似度，以及利用協同過濾方法預測使用者對電影的評分，我們選定一些電影，並將其電影海報以及簡介建立表單問卷並發送給我們的驗證目標，作為本研究的驗證方法，並將結果與前述的兩種推薦列表以及電影平均分數列表做比較，本研究經過驗證後得出類協同過濾推薦方法更能夠針對使用者進行精準的個人化推薦。

Previous research mainly focus on the same field of recommendation system, a few cross-domain studies use transfer learning method to do cross- domain recommendation, and most of them also establish networks between similar fields. Therefore, this study will try to design a cross domain recommendation system for social media, which will analyze the image post records of users in Instagram and build a heterogeneous network to connect with other field,in order to recommend information that might be interest to Instagram users in another field.
This research takes Instagram as the target domain, rotten tomato film review website as the source domain, and establish the cosine similarity recommendation method and the collaborative filtering recommendation method. We use the cosine similarity between Instagram users and movie reviewers, and use the collaborative filtering method to predict the user's rating of the movie. We select some films, uses the movie posters and movie introduction to build the questionnaire, and sent to our verification target users, as the verification method of this study, the results are compared with the above two recommendation lists and also compared with the movie average score list. this study concludes that the collaborative filtering recommendation method is more accurate and personalized for users.

目錄
中文摘要	I
英文摘要	II
目錄	IV
圖目錄	VI
表目錄	VII
第一章 緒論	1
1.1 背景與動機	1
1.2 研究目的	3
1.3 研究流程	3
第二章 文獻探討	5
2.1 推薦系統	5
2.2 網路表示學習法	8
2.3 整合異質資料相關研究	11
2.4 小結	12
第三章 研究方法	14
3.1 系統架構	14
3.2 模型建構	15
3.2.1 異質性網路設計	15
3.2.2 推薦模型	20
第四章 實驗結果	23
4.1 資料描述	23
4.2 實驗設定	26
4.3驗證方法	28
4.4 實驗結果分析與討論	34
第五章 結論與建議	40
5.1 結論	40
5.2 未來發展與建議	40
參考文獻	42
中文文獻	42
英文文獻	43

 
圖目錄
圖1-1 本研究流程圖	4
圖2-1 網絡表示法使用的示意圖	9
圖2-2 異質性網路連接示例	13
圖3-1 推薦系統架構圖	14
圖3-2 Google cloud Vision Web	16
圖3-3 Google cloud Vision Labels	17
圖3-4 爛番茄資料來源	18
圖3-5 異質性網路圖	19
圖3-6 node2vec 表示法	20
圖4-1 電影評分正規化	26
圖4-2 驗證問卷	29
圖4-3 推薦驗證機制	30
圖4-4 推薦途徑示例	37
圖4-5 使用者標籤示例	38
圖4-6 電影標籤	39

 
表目錄
表3-1 餘弦相似度推薦方法	21
表3-2 Instagram使用者與電影評論者餘弦相似度	22
表3-3 電影評分表	22
表4-1 Instagram資料結構	24
表4-2 爛番茄網站資料結構	25
表4-3 實驗資料的類型	27
表4-4 評估標準示例	31
表4-5 AP平均準確率（Average Precision）表現	34
表4-6 推薦列表示例	36

參考文獻
中文文獻
張伯新 .（2018）. 基於異質性資訊網路表示法學習之電子商務 推薦系統 . 國立政治大學資訊科學系碩士論文。
楊昇芳 .（2019）.基於超連結圖譜表示法之跨領域音樂推薦演算法. 國立政治大學資訊科學系碩士論文。
陳心萍 .（2017）. 學習基於異質性網路的領域感知網路表示法 . 台灣大學資訊網路與多媒體研究所學位論文。
 
英文文獻
Elior, Cohen（2018）. node2vec: Embeddings for Graph Data. Retrieved from: https://towardsdatascience.com/node2vec-embeddings-for-graph-data-32a866340fef
Adomavicius, G. and Tuzhilin, A.（ June 2005）. Towards the Next Generation of Recommender Systems: A Survey of the State-of-the-Art and Possible Extensions , IEEE Transactions on Knowledge and Data Engineering, pages : 734~749.
Bao, T., Ge, Y., Chen, E., Xiong, H. and Tian, J.（2012）. Collaborative filtering with user ratings and tags , Proc. 1st Int. Workshop Context Disc. Data Min., pages : 1~5. 
Carneiro, Nuno, Figueira, Gonçalo, Costa, Miguel（2017）. A data mining based system for credit-card fraud detection in e-tail , Decision Support Systems, ISSN: 0167-9236, Vol: 95, Page: 91~101.
Cui, Peng, Wang, Xiao, Pei, Jian, and Zhu0, Wenwu（May 1 2019）. A Survey on Network Embedding , IEEE Transactions on Knowledge and Data Engineering Volume: 31 , Issue: 5, Pages : 833~852
Grover and Leskovec, J.（2016）. Node2vec: Scalable feature learning for networks , In Proceedings of the 22Nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages : 855~864 

Hao, Peng, Zhang, Guangquan, Martinez, Luis, and Lu, Jie（Jan. 2019）. Regularizing Knowledge Transfer in Recommendation With Tag-Inferred Correlation , IEEE Transactions on Cybernetics Volume: 49 , Issue: 1, Pages : 83~96
He, Ruining, and McAuley, Julian（2016）. Ups and downs: Modeling the visual evolution of fashion trends with one-class collaborative filtering , In proceedings of the 25th international conference on world wide web, pages : 507~517.
He, Ruining, and McAuley, Julian（2016）. Vbpr: Visual bayesian personalized ranking from implicit feedback , AAAI Publications, Thirtieth AAAI Conference on Artificial Intelligence.
Hu, Y., Koren, Y. and Volinsky, C.（2008）. Collaborative filtering for implicit feedback datasets , Proceedings of the 2008 Eighth IEEE International Conference on Data Mining ICDM ’08, pages : 263~272 
Kim, Hyung W., Han, Keejun, Yi, Mun Y., Cho, Joonmyun, and Hong, Jinwoo（November 2012）. MovieMine: Personalized Movie Content Search by Utilizing User Comments , IEEE Transactions on Consumer Electronics Volume: 58 Issue: 4, Pages : 1416~1424.
Koren, Y., Bell, R. and Volinsky, C.（Aug 2009）. Matrix factorization techniques for recommender systems , Computer, pages : 30~37.
Liang, H., Xu, Y. , Li, Y., Nayak, R., and Tao, X.（2010）. Connecting users and items with weighted tags for personalized item recommendations , Proc. 21st ACM Conf. Hypertext Hypermedia, pages : 51~60. 
Ling, Guang, Michael R Lyu, and King, Irwin（2014）. Ratings meet reviews, a combined approach to recommend , In Proceedings of the 8th ACM Conference on Recommender systems, pages : 105-112.
Liu, Hongyan, He, Jun, Wang, Tingting, Song, Wenting, Du, Xiaoyang（2013）. Combining user preferences and user opinions for accurate recommendation , Electronic Commerce Research and Applications ISSN: 1567-4223, Vol: 12, Issue: 1, Pages : 14~23.
McAuley, Julian and Leskovec, Jure（2013）. Hidden factors and hidden topics: Understanding rating dimensions with review text , In Proceedings of the 7th ACM Conference on Recommender Systems, pages : 165~172.
McAuley, Julian, Targett, Christopher, Shi, Qinfeng and Anton Van Den Hengel（2015）. Image-based recommendations on styles and substitutes , In Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval, pages : 43~52.
Pan, Sinno Jialin, Yang, Qiang（Oct. 2010）. A Survey on Transfer Learning , IEEE Transactions on Knowledge and Data Engineering Volume: 22 , Issue: 10, Pages : 1345~ 1359.
 

Parra-Arnau, Javier, Perego, Andrea, Ferrari, Elena Fellow, IEEE, Forne ́, Jordi, and Rebollo-Monedero, David（Jan. 2014）. Privacy-Preserving Enhanced Collaborative Tagging , IEEE Transactions on Knowledge and Data Engineering Volume: 26 , Issue: 1, Pages : 180~193.
Ren, Jiangtao, Long, Jiawei, Xu, Zhikang（2019）. Financial news recommendation based on graph embeddings , Decision Support Systems, ISSN: 0167-9236, Vol: 125, Pages : 113115.
Shapira, B., Rokach, L., and Freilikhman , S.（2013）. Facebook single and cross domain data for recommendation systems, User Model. User Adapted Interact., vol. 23, nos. 2–3, pages : 211~247.
Shi, Chuan, Hu, Binbin, Zhao, Wayne Xin, and Yu, Philip S（2017）.Heterogeneous information network embedding for recommendation, IEEE Transactions on Knowledge and Data Engineering , pages : 357~370.
Shi, Y., Larson, M., and Hanjalic, A.（2011）. Tags as bridges between domains: Improving recommendation with tag-induced cross-domain collaborative filtering, Proc. 19th Int. Conf. User Model. Adaption Personalization, pages : 305~316.
Shi, Y., Larson, M., and Hanjalic, A.（May 2014）. Collaborative filtering beyond the user-item matrix: A survey of the state of the art and future challenges , ACM Comput. Surveys, vol. 47, no. 1, pages : 1~45.
Sun, Zhu, Guo, Qing, Yang, Jie, Fang, Hui, Guo, Guibing, Zhang, Jie Burke, Robin（2019）. Research commentary on recommendations with side information: A survey and research directions , Electronic Commerce Research and Applications, ISSN: 1567-4223, Vol: 37, Pages : 100879.
Wang, Quan, Mao, Zhendong, Wang, Bin, and Guo, Li（Dec. 1 2017）. Knowledge Graph Embedding: A Survey of Approaches and Applications , IEEE Transactions on Knowledge and Data Engineering Volume: 29 , Issue: 12, Pages : 2724~2743
Wu, Boya, Jia Jia, Yang Yang, Zhao, Peijun, Tang, Jie, and Tian, Qi（July 2017）. Inferring Emotional Tags From Social Images With User Demographics , IEEE Transactions on Multimedia Volume: 19 , Issue: 7, Pages : 1670~1684
Xin, X. et al., Cross-domain collaborative filtering with review text , Proc. 24th Int. Conf. Artif. Intell., Buenos Aires, Argentina, pages : 1827~1833.
Zhang, Wei, Yuan, Quan, Han, Jiawei, and Wang, Jianyong（2016）.Collaborative multi-level embedding learning from reviews for rating prediction. In Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence, IJCAI'16, pages : 2986~2992.