進階搜尋


下載電子全文  
系統識別號 U0026-3107201914015100
論文名稱(中文) 基於Aspect評論分析與深度學習之推薦系統
論文名稱(英文) A Recommendation System Using Aspect Analysis and Deep Learning.
校院名稱 成功大學
系所名稱(中) 資訊管理研究所
系所名稱(英) Institute of Information Management
學年度 107
學期 2
出版年 108
研究生(中文) 陳立欣
研究生(英文) Li-Hsin Chen
學號 R76061029
學位類別 碩士
語文別 中文
論文頁數 46頁
口試委員 指導教授-劉任修
口試委員-胡政宏
口試委員-張裕清
中文關鍵字 深度學習  矩陣分解  推薦系統  Attention網路 
英文關鍵字 Deep Learning  Matrix Factorization  Recommendation System  Attention 
學科別分類
中文摘要 近年來推薦系統已經應用在各行各業,傳統的推薦系統作法大多依賴用戶的歷史資訊或者物品特徵來進行推薦,而與情感分析結合的推薦系統也大多以整體的產品評價作為特徵依據,無法從細部資訊了解用戶對商品的各別喜好程度。
例如一則評論說明:「該手機的解析度很高,但效能很差」,其實體對象為手機,解析度和效能為Aspect,用戶對不同Aspect給出的評價可能不同。因此我們提出latent aspect的概念,將評論句子輸入深度學習模型中,並從句子中提取特徵向量,利用Attention機制將輸入的每個部分賦予不同的重要程度,分配較高的權重在用戶較為關注的Aspect評論描述上,進而提取關鍵的Aspect評論特徵,並結合機率矩陣分解法進行交互訓練,應用於推薦系統。我們使用Amazon review data公開資料集進行實驗,與(Kim et al., 2016)的卷積矩陣分解法方法進行比較,在三個資料集上,預測效果皆提升了2.5%以上。
英文摘要 In recent years, the recommendation system has been applied in all walks of life. Most of the traditional recommendation system practices rely on the user’s historical information or item characteristics to recommend a suitable choice. But the recommendation system combined with sentiment analysis usually uses the overall product evaluation as the feature basis, it is hard to understand the preference of individual products from the detailed information.
We propose the concept of latent aspect, input the comment sentence into the deep learning model, and extract the feature vector from the sentence. We Use the
Attention mechanism to assign each part of the input to different importance levels assign higher weights to the Aspect comment description of users’ attention. Our
goal is to extract key Aspect comment features, combining with a probability matrix decomposition method for interactive training, applied to the recommendation system to achieve better recommendation results.
論文目次 摘要i
EXTENDED ABSTRACT ii
誌謝x
目錄xi
表目錄xiv
圖目錄xv
1 緒論1
1.1 背景及動機. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 研究目的. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 貢獻. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4 論文架構. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2 相關文獻探討. . . . . . . . . . . . . . . . . . . . . . . . 7
2.1 深度學習方法. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.1 神經網路. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.2 Attention機制. . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.2 推薦系統方法. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.1 基於內容的推薦方法. . . . . . . . . . . . . . . . . . . . . . 10
2.2.2 協同過濾推薦方法. . . . . . . . . . . . . . . . . . . . . . . 10
2.2.3 混合式推薦方法. . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3 小結. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3 研究方法. . . . . . . . . . . . . . . . . . . . . . . . 14
3.1 問題描述. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.2 模型架構. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.3 模型架構與訓練方法. . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.3.1 Word Encoder . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.3.2 Word Attention . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.3.3 Sentence Encoder . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.3.4 Sentence Attention for Aspect . . . . . . . . . . . . . . . . . . 21
3.3.5 Sentence Attention for Document . . . . . . . . . . . . . . . . 21
3.3.6 Sentiment Embedding . . . . . . . . . . . . . . . . . . . . . . 22
3.3.7 Fully Connected . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.3.8 Loss function . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.3.9 HAN模型結合機率矩陣分解法. . . . . . . . . . . . . . . . . 23
4 實驗與分析. . . . . . . . . . . . . . . . . . . . . . . . 31
4.1 實驗架構及步驟. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4.2 資料集與資料處理. . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.3 實驗結果與分析. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.3.1 衡量指標. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.3.2 實驗環境與參數設定. . . . . . . . . . . . . . . . . . . . . . 35
4.3.3 實驗結果與分析. . . . . . . . . . . . . . . . . . . . . . . . . 38
5 結論與未來發展. . . . . . . . . . . . . . . . . . . . . . . . 42
參考文獻. . . . . . . . . . . . . . . . . . . . . . . . 43
參考文獻 Ahn, H. J. (2008). A new similarity measure for collaborative filtering to alleviate the new user cold-starting problem. Information Sciences, 178(1):37–51.
Bahdanau, D., Cho, K., and Bengio, Y. (2014). Neural machine translation by jointly learning to align and translate. arXiv preprint arXiv:1409.0473.
Balabanovic´, M. and Shoham, Y. (1997). Fab: content-based, collaborative recommen- dation. Communications of the ACM, 40(3):66–72.
Bickart, B. and Schindler, R. M. (2001). Internet forums as influential sources of con- sumer information. Journal of interactive marketing, 15(3):31–40.
Cho, K., Van Merrie¨nboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., and Bengio, Y. (2014). Learning phrase representations using rnn encoder-decoder for statistical machine translation. arXiv preprint arXiv:1406.1078.
Chung, J., Gulcehre, C., Cho, K., and Bengio, Y. (2014). Empirical evaluation of gated recurrent neural networks on sequence modeling. arXiv preprint arXiv:1412.3555.
Davidson, J., Liebald, B., Liu, J., Nandy, P., Van Vleet, T., Gargi, U., Gupta, S., He, Y., Lambert, M., Livingston, B., et al. (2010). The youtube video recommendation system. In Proceedings of the fourth ACM conference on Recommender systems, pages 293–296. ACM.
Hochreiter, S. and Schmidhuber, J. (1997). Long short-term memory. Neural computa- tion, 9(8):1735–1780.
Karypis, G. (2001). Evaluation of item-based top-n recommendation algorithms. In Proceedings of the tenth international conference on Information and knowledge management, pages 247–254. ACM.
Kim, D., Park, C., Oh, J., Lee, S., and Yu, H. (2016). Convolutional matrix factoriza- tion for document context-aware recommendation. In Proceedings of the 10th ACM Conference on Recommender Systems, pages 233–240. ACM.
Lei, X., Qian, X., and Zhao, G. (2016). Rating prediction based on social sentiment from textual reviews. IEEE Transactions on Multimedia, 18(9):1910–1921.
Maas, A. L., Daly, R. E., Pham, P. T., Huang, D., Ng, A. Y., and Potts, C. (2011). Learn- ing word vectors for sentiment analysis. In Proceedings of the 49th annual meeting of the association for computational linguistics: Human language technologies-volume 1, pages 142–150. Association for Computational Linguistics.
Mandl, M., Felfernig, A., Teppan, E., and Schubert, M. (2011). Consumer decision making in knowledge-based recommendation. Journal of Intelligent Information Sys- tems, 37(1):1–22.
McAuley, J., Pandey, R., and Leskovec, J. (2015). Inferring networks of substitutable and complementary products. In Proceedings of the 21th ACM SIGKDD interna- tional conference on knowledge discovery and data mining, pages 785–794. ACM.
Mikolov, T., Karafia´t, M., Burget, L., Cˇ ernocky`, J., and Khudanpur, S. (2010). Recur- rent neural network based language model. In Eleventh Annual Conference of the International Speech Communication Association.
Mnih, A. and Salakhutdinov, R. R. (2008). Probabilistic matrix factorization. In Ad- vances in neural information processing systems, pages 1257–1264.
Pazzani, M. J. and Billsus, D. (2007). Content-based recommendation systems. In The adaptive web, pages 325–341. Springer.
Ratner, A., Hancock, B., Dunnmon, J., Sala, F., Pandey, S., and Re´, C. (2018). Training complex models with multi-task weak supervision. arXiv preprint arXiv:1810.02840.
Rush, A. M., Chopra, S., and Weston, J. (2015). A neural attention model for abstractive sentence summarization. arXiv preprint arXiv:1509.00685.
Sundermeyer, M., Schlu¨ter, R., and Ney, H. (2012). Lstm neural networks for language modeling. In Thirteenth annual conference of the international speech communica- tion association.
Wang, C. and Blei, D. M. (2011). Collaborative topic modeling for recommending scientific articles. In Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 448–456. ACM.
Wang, Y., Huang, M., Zhao, L., et al. (2016). Attention-based lstm for aspect-level sentiment classification. In Proceedings of the 2016 conference on empirical methods in natural language processing, pages 606–615.
Yang, Z., Yang, D., Dyer, C., He, X., Smola, A., and Hovy, E. (2016). Hierarchical at- tention networks for document classification. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 1480–1489.

Yi, B., Shen, X., Zhang, Z., Shu, J., and Liu, H. (2016). Expanded autoencoder recom- mendation framework and its application in movie recommendation. In 2016 10th International Conference on Software, Knowledge, Information Management & Ap- plications (SKIMA), pages 298–303. IEEE.
Zeng, C., Xing, C.-X., Zhou, L.-Z., and Zheng, X.-H. (2004). Similarity measure and in- stance selection for collaborative filtering. International Journal of Electronic Com- merce, 8(4):115–129.
Zhang, S., Wang, W., Ford, J., Makedon, F., and Pearlman, J. (2005). Using singular value decomposition approximation for collaborative filtering. In Seventh IEEE Inter- national Conference on E-Commerce Technology (CEC’05), pages 257–264. IEEE.
Zhang, Y., Lai, G., Zhang, M., Zhang, Y., Liu, Y., and Ma, S. (2014). Explicit factor models for explainable recommendation based on phrase-level sentiment analysis. In Proceedings of the 37th international ACM SIGIR conference on Research & devel- opment in information retrieval, pages 83–92. ACM.
Zheng, L., Noroozi, V., and Yu, P. S. (2017). Joint deep modeling of users and items using reviews for recommendation. In Proceedings of the Tenth ACM International Conference on Web Search and Data Mining, pages 425–434. ACM.
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2019-08-06起公開。
  • 同意授權校外瀏覽/列印電子全文服務,於2019-08-06起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw