Publications-Theses
Article View/Open
Publication Export
-
題名 基於時序與風格的語音節目推薦系統研究
An investigation of spoken program recommendation systems based on time and style作者 蘇品維
Su, Pin-Wei貢獻者 杜雨儒
Tu,Yu-Ju
蘇品維
Su,Pin-Wei關鍵詞 推薦系統
冷啟動問題
Podcast
機器學習
時間
敘事風格
: Recommendation systems
Cold-start problem
Podcast
Machine Learning
Listening Time
Speaking Style日期 2022 上傳時間 1-Aug-2022 17:22:56 (UTC+8) 摘要 因為網際網路以及串流技術的蓬勃發展，在相關的市場上開始衍伸出許多新媒體服務，例如:影音串流(YouTube) 語音節目(Podcast)。其中又因為語音節目的一些自身的特性，以及在台灣市場當中發生了在短時間內湧入大量新用戶以及新產品的問題，使得在語音節目平台中建置推薦系統顯得相當困難，特別是在推薦新產品或是推薦產品給新用戶的議題上。換句話說，要如何推薦適合的語音節目給新的使用者 或是如何推薦新的語音節目給使用者等議題顯得十分重要且具有挑戰性。這些議題在過去的研究中這些問題也被稱作「冷啟動問題」。\n而過去對於語音節目推薦上的相關研究十分稀少。而本研究基於在過去少量的語音節目推薦文獻上所使用之文字描述和類別的特性，結合兩個特有的特性—「時序」與語音節目的「敘事風格」，並搭配混合推薦的方法，去解決在語音節目上所發生的冷啟動問題。本研究也透過網路爬蟲的方式蒐集 APPLE Podcast 的評分資料作為推薦系統的訓練資料。除了線下測試外，也實作出推薦系統介面實際讓使用者進行使用者測試。根據我們的線下測試以及使用者測試之結果可以得知在時序與風格屬性的幫助下，在整體以及新使用者的推薦效果都有顯著的提升。
With the rapid development of the Internet and streaming technology, many new media services have begun to spread to related markets, such as video streaming (YouTube) and spoken program services (podcasts). Due to the characteristics of spoken programs and the influx of new users and products in the Taiwan market in a short time, it is challenging to build a recommendation system in the spoken program platform, especially on the issues of recommending new items or recommending items to new users. In other words, the challenges of recommending suitable spoken programs to new users or recommending new spoken programs to users remain open research questions. These issues are called “cold-start problems” in past studies. Related research on spoken program recommendations was scarce in the past. The present study, based on the characteristics of text descriptions and categories used in a few recommended literatures of spoken programs in the past, combined two unique characteristics—listening time and speaking styles of spoken programs—in a hybrid recommendation method to solve the cold-start problems in spoken programs. The study also collected the rating data of Apple Podcasts through the web crawler, and podcast ratings were used as training data for the recommendation system. In addition to offline testing, we implemented a recommended system interface for users to conduct user testing. According to the results of our offline and user tests, with the addition of time and speaking styles, the performance of the recommendation in overall and new users was significantly improved.參考文獻 1. An, H. S. 2017. "Breaking the Pop Music Market in the Us: Effects of Online PrePurchase Influences and Acculturation of Asian Consumers on Their Responses\nto Asian Artists’ Contemporary Pop Music," Proceedings of the Northeast\nBusiness & Economics Association).\n2. Arif, I., Aslam, W., and Siddiqui, H. 2020. "Influence of Brand Related UserGenerated Content through Facebook on Consumer Behaviour: A StimulusOrganism-Response Framework," International Journal of Electronic Business\n(15:2), pp. 109-132.\n3. Ayata, D., Yaslan, Y., and Kamasak, M. E. 2018. "Emotion Based Music\nRecommendation System Using Wearable Physiological Sensors," IEEE\ntransactions on consumer electronics (64:2), pp. 196-203.\n4. Bai, T., Wen, J.-R., Zhang, J., and Zhao, W. X. 2017. "A Neural Collaborative\nFiltering Model with Interaction-Based Neighborhood," Proceedings of the 2017\nACM on Conference on Information and Knowledge Management, pp. 1979-1982.\n5. Balabanović, M., and Shoham, Y. 1997. "Fab: Content-Based, Collaborative\nRecommendation," Commun. ACM (40:3), pp. 66–72.\n6. Beel, J., Langer, S., Genzmehr, M., and Nürnberger, A. 2013. "Introducing\nDocear`s Research Paper Recommender System," Proceedings of the 13th\nACM/IEEE-CS joint conference on Digital libraries, pp. 459-460.\n7. Belk, R. W. 1974. "An Exploratory Assessment of Situational Effects in Buyer\nBehavior," Journal of marketing research (11:2), pp. 156-163.\n8. Benton, G., Fazelnia, G., Wang, A., and Carterette, B. 2020. "Trajectory Based\nPodcast Recommendation," arXiv preprint arXiv:2009.03859).\n9. Bobadilla, J., Ortega, F., Hernando, A., and Bernal, J. 2012. "A Collaborative\nFiltering Approach to Mitigate the New User Cold Start Problem," Knowledgebased systems (26), pp. 225-238.\n10. Burke, R. 2002. "Hybrid Recommender Systems: Survey and Experiments," User\nmodeling and user-adapted interaction (12:4), pp. 331-370.\n11. Cai, J.-J., Tang, J., Chen, Q.-G., Hu, Y., Wang, X., and Huang, S.-J. 2019. "MultiView Active Learning for Video Recommendation," IJCAI, pp. 2053-2059.\n12. Chen, C.-M., and Sun, Y.-C. 2012. "Assessing the Effects of Different Multimedia\nMaterials on Emotions and Learning Performance for Visual and Verbal Style\nLearners," Computers & Education (59:4), pp. 1273-1285.\n13. Chen, H.-H., Chung, C.-A., Huang, H.-C., and Tsui, W. 2017. "Common Pitfalls\nin Training and Evaluating Recommender Systems," ACM SIGKDD Explorations\nNewsletter (19:1), pp. 37-45.\n14. Choi, S.-M., Jang, K., Lee, T.-D., Khreishah, A., and Noh, W. 2020. "Alleviating\nItem-Side Cold-Start Problems in Recommender Systems Using Weak\nSupervision," IEEE Access (8), pp. 167747-167756.\n15. Collins, A., Tkaczyk, D., Aizawa, A., and Beel, J. 2018. "Position Bias in\nRecommender Systems for Digital Libraries," International Conference on\nInformation: Springer, pp. 335-344.\n16. Common-Wealth-Magazine ,2021. Complete investigation report of "2021 Listening to\nEconomic Survey" https://www.cw.com.tw/article/5115003?template=transformers (Last\naccess date 2022/07)\n77\n17. Darshna, P. 2018. "Music Recommendation Based on Content and Collaborative\nApproach & Reducing Cold Start Problem," 2018 2nd International Conference\non Inventive Systems and Control (ICISC): IEEE, pp. 1033-1037.\n18. Das, D., Sahoo, L., and Datta, S. 2017. "A Survey on Recommendation System,"\nInternational Journal of Computer Applications (160:7).\n19. Deng, S., Wang, D., Li, X., and Xu, G. 2015. "Exploring User Emotion in\nMicroblogs for Music Recommendation," Expert Systems with Applications\n(42:23), pp. 9284-9293.\n20. Díaz-Morales, J. F., Escribano, C., and Jankowski, K. S. 2015. "Chronotype and\nTime-of-Day Effects on Mood During School Day," Chronobiology international\n(32:1), pp. 37-42.\n21. Felício, C. Z., Paixão, K. V., Barcelos, C. A., and Preux, P. 2017. "A Multi-Armed\nBandit Model Selection for Cold-Start User Recommendation," Proceedings of\nthe 25th Conference on User Modeling, Adaptation and Personalization, pp. 32-\n40.\n22. Gunawan, A. A., and Suhartono, D. 2019. "Music Recommender System Based\non Genre Using Convolutional Recurrent Neural Networks," Procedia Computer\nScience (157), pp. 99-109.\n23. Guo, G., Zhang, J., and Yorke-Smith, N. 2015. "Trustsvd: Collaborative Filtering\nwith Both the Explicit and Implicit Influence of User Trust and of Item Ratings,"\nProceedings of the AAAI Conference on Artificial Intelligence.\n24. He, X., Du, X., Wang, X., Tian, F., Tang, J., and Chua, T.-S. 2018. "Outer ProductBased Neural Collaborative Filtering," arXiv preprint arXiv:1808.03912).\n25. He, X., Liao, L., Zhang, H., Nie, L., Hu, X., and Chua, T.-S. 2017. "Neural\nCollaborative Filtering," Proceedings of the 26th international conference on\nworld wide web, pp. 173-182.\n26. Herce-Zelaya, J., Porcel, C., Bernabé-Moreno, J., Tejeda-Lorente, A., and\nHerrera-Viedma, E. 2020. "New Technique to Alleviate the Cold Start Problem in\nRecommender Systems Using Information from Social Media and Random\nDecision Forests," Information Sciences (536), pp. 156-170.\n27. Herlocker, J. L., Konstan, J. A., Borchers, A., and Riedl, J. 1999. "An Algorithmic\nFramework for Performing Collaborative Filtering," Proceedings of the 22nd\nannual international ACM SIGIR conference on Research and development in\ninformation retrieval, pp. 230-237.\n28. Hidalgo, M. P., Caumo, W., Posser, M., Coccaro, S. B., Camozzato, A. L., and\nChaves, M. L. F. 2009. "Relationship between Depressive Mood and Chronotype\nin Healthy Subjects," Psychiatry and clinical neurosciences (63:3), pp. 283-290.\n29. Hu, Y., and Ogihara, M. 2011. "Nextone Player: A Music Recommendation\nSystem Based on User Behavior," ISMIR, pp. 103-108.\n30. Hyung, Z., Lee, K., and Lee, K. 2014. "Music Recommendation Using Text\nAnalysis on Song Requests to Radio Stations," Expert Systems with Applications\n(41:5), pp. 2608-2618.\n31. Inside ,2020. Show you Podcast- 2020 Taiwan Podcast Industry Report for the\nFirst Half of the Year https://www.inside.com.tw/article/20391-2020-podcastreport (Last access date 2022/07)\n32. Ismailoglu, F. 2021. "Aggregating User Preferences in Group Recommender\nSystems: A Crowdsourcing Approach," Decision Support Systems), p. 113663.\n33. Jelodar, H., Wang, Y., Rabbani, M., Ahmadi, S. B. B., Boukela, L., Zhao, R., and\nLarik, R. S. A. 2021. "A Nlp Framework Based on Meaningful Latent-Topic\n78\nDetection and Sentiment Analysis Via Fuzzy Lattice Reasoning on Youtube\nComments," Multimedia Tools and Applications (80:3), pp. 4155-4181.\n34. Joo, Y. S., and Kim, S.-K. 2020. "Scent Emotion Evaluation Experiment for\nMultimedia Application," 2020 IEEE International Conference on Consumer\nElectronics-Asia (ICCE-Asia): IEEE, pp. 1-4.\n35. Kaji, N., and Kobayashi, H. 2017. "Incremental Skip-Gram Model with Negative\nSampling," arXiv preprint arXiv:1704.03956).\n36. Kim, H.-N., Ha, I., Lee, K.-S., Jo, G.-S., and El-Saddik, A. 2011. "Collaborative\nUser Modeling for Enhanced Content Filtering in Recommender Systems,"\nDecision Support Systems (51:4), pp. 772-781.\n37. Kudielka, B. M., Schommer, N. C., Hellhammer, D. H., and Kirschbaum, C. 2004.\n"Acute Hpa Axis Responses, Heart Rate, and Mood Changes to Psychosocial\nStress (Tsst) in Humans at Different Times of Day," Psychoneuroendocrinology\n(29:8), pp. 983-992.\n38. Kumar, P., and Thakur, R. S. 2018. "Recommendation System Techniques and\nRelated Issues: A Survey," International Journal of Information Technology\n(10:4), pp. 495-501.\n39. Lang, A., and Chrzan, J. 2015. "Media Multitasking: Good, Bad, or Ugly?,"\nAnnals of the International Communication Association (39:1), pp. 99-128.\n40. Lee, J.-S., and Shin, D.-H. 2015. "A Study on the Preference between Emotion of\nHuman and Media Genre in Smart Device," Science of Emotion and Sensibility\n(18:1), pp. 59-66.\n41. Li, G., and Zhang, J. 2018. "Music Personalized Recommendation System Based\non Improved Knn Algorithm," 2018 IEEE 3rd Advanced Information Technology,\nElectronic and Automation Control Conference (IAEAC): IEEE, pp. 777-781.\n42. Li, Y.-M., Liou, J.-H., and Li, Y.-W. 2020. "A Social Recommendation Approach\nfor Reward-Based Crowdfunding Campaigns," Information & Management (57:7),\np. 103246.\n43. Liang, T.-P., Lai, H.-J., and Ku, Y.-C. 2006. "Personalized Content\nRecommendation and User Satisfaction: Theoretical Synthesis and Empirical\nFindings," Journal of Management Information Systems (23:3), pp. 45-70.\n44. Llisterri, J. 1992. "Speaking Styles in Speech Research," Workshop on Integrating\nSpeech and Natural Language: Citeseer.\n45. Marshall, D., and Sidorov, K. 2001. "Introduction to Multimedia," UK: School of\nComputer Science & Informatics Cardiff University).\n46. Martikainen, K. 2020. "Audio-Based Stylistic Characteristics of Podcasts for\nSearch and Recommendation: A User and Computational Analysis." University of\nTwente.\n47. Massquantity, 2021. LibRecommender [Source code].\nhttps://github.com/massquantity/LibRecommender. (Last access date 2022/07)\n48. Mehrabian, A., and Russell, J. A. 1974. An Approach to Environmental\nPsychology. the MIT Press.\n49. Moscato, V., Picariello, A., and Sperli, G. 2020. "An Emotional Recommender\nSystem for Music," IEEE Intelligent Systems).\n50. Murray, G., Nicholas, C. L., Kleiman, J., Dwyer, R., Carrington, M. J., Allen, N.\nB., and Trinder, J. 2009. "Nature’s Clocks and Human Mood: The Circadian\nSystem Modulates Reward Motivation," Emotion (9:5), p. 705.\n79\n51. National Audiovisual Institute, 2022.[Source code]\nhttps://github.com/ina-foss/inaSpeechSegmenter (Last access date 2022/07)\n52. Nazari, Z., Charbuillet, C., Pages, J., Laurent, M., Charrier, D., Vecchione, B., and\nCarterette, B. 2020. "Recommending Podcasts for Cold-Start Users Based on\nMusic Listening and Taste," in: Proceedings of the 43rd International ACM SIGIR\nConference on Research and Development in Information Retrieval. Virtual Event,\nChina: Association for Computing Machinery, pp. 1041–1050.\n53. Panagiotakis, C., Papadakis, H., and Fragopoulou, P. 2021. "A Dual Hybrid\nRecommender System Based on Scor and the Random Forest," Computer Science\nand Information Systems (18:1), pp. 115-128.\n54. Raheem, K. R., and Ali, I. H. 2020. "Multimodal Content-Based Recommender\nSystem Using Three-Dimension Convolution Neural Network," COMPUSOFT:\nAn International Journal of Advanced Computer Technology (9:5).\n55. Rana, M. K. C. 2012. "Survey Paper on Recommendation System,").\n56. Rendle, S. 2010. "Factorization Machines," 2010 IEEE International conference\non data mining: IEEE, pp. 995-1000.\n57. Roenneberg, T. 2012. "What Is Chronotype?," Sleep and biological rhythms (10:2),\npp. 75-76.\n58. Ruff, J. 2002. "Information Overload: Causes, Symptoms and Solutions," Harvard\nGraduate School of Education), pp. 1-13.\n59. Ryu, J., Capistrano, E. P., and Lin, H.-C. 2020. "Non-Korean Consumers’\nPreferences on Korean Popular Music: A Two-Country Study," International\nJournal of Market Research (62:2), pp. 234-252.\n60. Sanz-Cruzado, J., Castells, P., and López, E. 2019. "A Simple Multi-Armed\nNearest-Neighbor Bandit for Interactive Recommendation," Proceedings of the\n13th ACM Conference on Recommender Systems, pp. 358-362.\n61. Sarwar, B., Karypis, G., Konstan, J., and Riedl, J. 2000. "Application of\nDimensionality Reduction in Recommender System-a Case Study," Minnesota\nUniv Minneapolis Dept of Computer Science.\n62. Selvi, C., and Sivasankar, E. 2019. "An Efficient Context-Aware Music\nRecommendation Based on Emotion and Time Context," in Data Science and Big\nData Analytics. Springer, pp. 215-228.\n63. Shani, G., and Gunawardana, A. 2011. "Evaluating Recommendation Systems,"\nin Recommender Systems Handbook. Springer, pp. 257-297.\n64. Son, J., and Kim, S. B. 2018. "Academic Paper Recommender System Using\nMultilevel Simultaneous Citation Networks," Decision Support Systems (105), pp.\n24-33.\n65. Subramaniyaswamy, V., and Logesh, R. 2017. "Adaptive Knn Based\nRecommender System through Mining of User Preferences," Wireless Personal\nCommunications (97:2), pp. 2229-2247.\n66. Sulaiman, N., Muhammad, A. M., Ganapathy, N. N. D. F., Khairuddin, Z., and\nOthman, S. 2017. "A Comparison of Students` Performances Using Audio Only\nand Video Media Methods," English language teaching (10:7), pp. 210-215.\n67. Taylor, B. J., and Hasler, B. P. 2018. "Chronotype and Mental Health: Recent\nAdvances," Current psychiatry reports (20:8), pp. 1-10.\n68. Thales,2021. PySpeech [Source code]\nhttps://github.com/thalesaguiar21/PySpeech (Last access date 2022/07)\n69. Thorat, P. B., Goudar, R., and Barve, S. 2015. "Survey on Collaborative Filtering,\nContent-Based Filtering and Hybrid Recommendation System," International\nJournal of Computer Applications (110:4), pp. 31-36.\n80\n70. Tsagkias, M., Larson, M., and De Rijke, M. 2010. "Predicting Podcast Preference:\nAn Analysis Framework and Its Application," Journal of the American Society for\ninformation Science and Technology (61:2), pp. 374-391.\n71. Volkovs, M., Yu, G. W., and Poutanen, T. 2017. "Content-Based Neighbor\nModels for Cold Start in Recommender Systems," in Proceedings of the\nRecommender Systems Challenge 2017. pp. 1-6.\n72. Volokhin, S., and Agichtein, E. 2018. "Understanding Music Listening Intents\nDuring Daily Activities with Implications for Contextual Music\nRecommendation," Proceedings of the 2018 Conference on Human Information\nInteraction & Retrieval, pp. 313-316.\n73. Wang, D., Zhang, X., Yu, D., Xu, G., and Deng, S. 2020a. "Came: Content-and\nContext-Aware Music Embedding for Recommendation," IEEE transactions on\nneural networks and learning systems (32:3), pp. 1375-1388.\n74. Wang, R., Ma, X., Jiang, C., Ye, Y., and Zhang, Y. 2020b. "Heterogeneous\nInformation Network-Based Music Recommendation System in Mobile\nNetworks," Computer Communications (150), pp. 429-437.\n75. Wang, X., Rosenblum, D., and Wang, Y. 2012. "Context-Aware Mobile Music\nRecommendation for Daily Activities," Proceedings of the 20th ACM\ninternational conference on Multimedia, pp. 99-108.\n76. Xing, Z., Parandehgheibi, M., Xiao, F., Kulkarni, N., and Pouliot, C. 2016.\n"Content-Based Recommendation for Podcast Audio-Items Using Natural\nLanguage Processing Techniques," 2016 IEEE International Conference on Big\nData (Big Data): IEEE, pp. 2378-2383.\n77. Xu, L., Wen, X., Shi, J., Li, S., Xiao, Y., Wan, Q., and Qian, X. 2021. "Effects of\nIndividual Factors on Perceived Emotion and Felt Emotion of Music: Based on\nMachine Learning Methods," Psychology of Music (49:5), pp. 1069-1087.\n78. Xu, X., Dutta, K., and Ge, C. 2018. "Do Adjective Features from User Reviews\nAddress Sparsity and Transparency in Recommender Systems?," Electronic\nCommerce Research and Applications (29), pp. 113-123.\n79. Yang, L., Sobolev, M., Wang, Y., Chen, J., Dunne, D., Tsangouri, C., Dell, N.,\nNaaman, M., and Estrin, D. 2019. "How Intention Informed Recommendations\nModulate Choices: A Field Study of Spoken Word Content," The World Wide Web\nConference, pp. 2169-2180.\n80. Ye, B. K., Tu, Y. J. T., and Liang, T. P. 2019. "A Hybrid System for Personalized\nContent Recommendation," Journal of Electronic Commerce Research (20:2), pp.\n91-104.\n81. Yin, H., Cui, B., Li, J., Yao, J., and Chen, C. 2012. "Challenging the Long Tail\nRecommendation," arXiv preprint arXiv:1205.6700).\n82. Yu, T., Guo, J., Li, W., and Lu, M. 2021. "A Mixed Heterogeneous Factorization\nModel for Non-Overlapping Cross-Domain Recommendation," Decision Support\nSystems), p. 113625.\n83. Zhang, H., Lu, Y., Gupta, S., and Zhao, L. 2014. "What Motivates Customers to\nParticipate in Social Commerce? The Impact of Technological Environments and\nVirtual Customer Experiences," Information & Management (51:8), pp. 1017-\n1030.\n84. Zhu, Y., Lin, J., He, S., Wang, B., Guan, Z., Liu, H., and Cai, D. 2019. "Addressing\nthe Item Cold-Start Problem by Attribute-Driven Active Learning," IEEE\nTransactions on Knowledge and Data Engineering (32:4), pp. 631-644 描述 碩士
國立政治大學
資訊管理學系
109356024資料來源 http://thesis.lib.nccu.edu.tw/record/#G0109356024 資料類型 thesis dc.contributor.advisor 杜雨儒 zh_TW dc.contributor.advisor Tu,Yu-Ju en_US dc.contributor.author (Authors) 蘇品維 zh_TW dc.contributor.author (Authors) Su,Pin-Wei en_US dc.creator (作者) 蘇品維 zh_TW dc.creator (作者) Su, Pin-Wei en_US dc.date (日期) 2022 en_US dc.date.accessioned 1-Aug-2022 17:22:56 (UTC+8) - dc.date.available 1-Aug-2022 17:22:56 (UTC+8) - dc.date.issued (上傳時間) 1-Aug-2022 17:22:56 (UTC+8) - dc.identifier (Other Identifiers) G0109356024 en_US dc.identifier.uri (URI) https://ah.lib.nccu.edu.tw/item?item_id=160398 - dc.description (描述) 碩士 zh_TW dc.description (描述) 國立政治大學 zh_TW dc.description (描述) 資訊管理學系 zh_TW dc.description (描述) 109356024 zh_TW dc.description.abstract (摘要) 因為網際網路以及串流技術的蓬勃發展，在相關的市場上開始衍伸出許多新媒體服務，例如:影音串流(YouTube) 語音節目(Podcast)。其中又因為語音節目的一些自身的特性，以及在台灣市場當中發生了在短時間內湧入大量新用戶以及新產品的問題，使得在語音節目平台中建置推薦系統顯得相當困難，特別是在推薦新產品或是推薦產品給新用戶的議題上。換句話說，要如何推薦適合的語音節目給新的使用者 或是如何推薦新的語音節目給使用者等議題顯得十分重要且具有挑戰性。這些議題在過去的研究中這些問題也被稱作「冷啟動問題」。\n而過去對於語音節目推薦上的相關研究十分稀少。而本研究基於在過去少量的語音節目推薦文獻上所使用之文字描述和類別的特性，結合兩個特有的特性—「時序」與語音節目的「敘事風格」，並搭配混合推薦的方法，去解決在語音節目上所發生的冷啟動問題。本研究也透過網路爬蟲的方式蒐集 APPLE Podcast 的評分資料作為推薦系統的訓練資料。除了線下測試外，也實作出推薦系統介面實際讓使用者進行使用者測試。根據我們的線下測試以及使用者測試之結果可以得知在時序與風格屬性的幫助下，在整體以及新使用者的推薦效果都有顯著的提升。 zh_TW dc.description.abstract (摘要) With the rapid development of the Internet and streaming technology, many new media services have begun to spread to related markets, such as video streaming (YouTube) and spoken program services (podcasts). Due to the characteristics of spoken programs and the influx of new users and products in the Taiwan market in a short time, it is challenging to build a recommendation system in the spoken program platform, especially on the issues of recommending new items or recommending items to new users. In other words, the challenges of recommending suitable spoken programs to new users or recommending new spoken programs to users remain open research questions. These issues are called “cold-start problems” in past studies. Related research on spoken program recommendations was scarce in the past. The present study, based on the characteristics of text descriptions and categories used in a few recommended literatures of spoken programs in the past, combined two unique characteristics—listening time and speaking styles of spoken programs—in a hybrid recommendation method to solve the cold-start problems in spoken programs. The study also collected the rating data of Apple Podcasts through the web crawler, and podcast ratings were used as training data for the recommendation system. In addition to offline testing, we implemented a recommended system interface for users to conduct user testing. According to the results of our offline and user tests, with the addition of time and speaking styles, the performance of the recommendation in overall and new users was significantly improved. en_US dc.description.tableofcontents Table of Contents\nCHAPTER 1: Introduction 6\n1-1 Research motivation 6\nCHAPTER2: Literature Review 10\n2-1 Methods in recommendation systems 10\n2-2 New user and new items recommendations 13\n2-3 Related works in hybrid multimedia recommendation systems 14\n2-3-1 Hybrid method in multimedia recommendations 14\n2-3-2 The features of user and item in multimedia recommendations 22\n2-4 The features of spoken program recommendations 24\n2-4-1 Text-based recommendation 24\n2-4-2 Speaking style 27\n2-4-3 Listening Time 28\nCHAPTER3: Proposed Model 31\nCHAPTER4: Empirical Experiments 39\n4-1 Baselines 39\n4-2 Experimental design 39\n4-2-1 The offline test approach 40\n4-2-2 The user test approach 47\nCHAPTER 5: Summary of Findings and Discussion 55\n5-1 The findings in offline test 55\n5-2 The findings in user test 61\n5-2-1 Data description of the user test samples 61\n5-2-2 The findings in user test 63\n5-3 Discussion 70\nCHAPTER6: Conclusion 74\nReferences 76\nAppendix 81 zh_TW dc.format.extent 3087016 bytes - dc.format.mimetype application/pdf - dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0109356024 en_US dc.subject (關鍵詞) 推薦系統 zh_TW dc.subject (關鍵詞) 冷啟動問題 zh_TW dc.subject (關鍵詞) Podcast zh_TW dc.subject (關鍵詞) 機器學習 zh_TW dc.subject (關鍵詞) 時間 zh_TW dc.subject (關鍵詞) 敘事風格 zh_TW dc.subject (關鍵詞) : Recommendation systems en_US dc.subject (關鍵詞) Cold-start problem en_US dc.subject (關鍵詞) Podcast en_US dc.subject (關鍵詞) Machine Learning en_US dc.subject (關鍵詞) Listening Time en_US dc.subject (關鍵詞) Speaking Style en_US dc.title (題名) 基於時序與風格的語音節目推薦系統研究 zh_TW dc.title (題名) An investigation of spoken program recommendation systems based on time and style en_US dc.type (資料類型) thesis en_US dc.relation.reference (參考文獻) 1. An, H. S. 2017. "Breaking the Pop Music Market in the Us: Effects of Online PrePurchase Influences and Acculturation of Asian Consumers on Their Responses\nto Asian Artists’ Contemporary Pop Music," Proceedings of the Northeast\nBusiness & Economics Association).\n2. Arif, I., Aslam, W., and Siddiqui, H. 2020. "Influence of Brand Related UserGenerated Content through Facebook on Consumer Behaviour: A StimulusOrganism-Response Framework," International Journal of Electronic Business\n(15:2), pp. 109-132.\n3. Ayata, D., Yaslan, Y., and Kamasak, M. E. 2018. "Emotion Based Music\nRecommendation System Using Wearable Physiological Sensors," IEEE\ntransactions on consumer electronics (64:2), pp. 196-203.\n4. Bai, T., Wen, J.-R., Zhang, J., and Zhao, W. X. 2017. "A Neural Collaborative\nFiltering Model with Interaction-Based Neighborhood," Proceedings of the 2017\nACM on Conference on Information and Knowledge Management, pp. 1979-1982.\n5. Balabanović, M., and Shoham, Y. 1997. "Fab: Content-Based, Collaborative\nRecommendation," Commun. ACM (40:3), pp. 66–72.\n6. Beel, J., Langer, S., Genzmehr, M., and Nürnberger, A. 2013. "Introducing\nDocear`s Research Paper Recommender System," Proceedings of the 13th\nACM/IEEE-CS joint conference on Digital libraries, pp. 459-460.\n7. Belk, R. W. 1974. "An Exploratory Assessment of Situational Effects in Buyer\nBehavior," Journal of marketing research (11:2), pp. 156-163.\n8. Benton, G., Fazelnia, G., Wang, A., and Carterette, B. 2020. "Trajectory Based\nPodcast Recommendation," arXiv preprint arXiv:2009.03859).\n9. Bobadilla, J., Ortega, F., Hernando, A., and Bernal, J. 2012. "A Collaborative\nFiltering Approach to Mitigate the New User Cold Start Problem," Knowledgebased systems (26), pp. 225-238.\n10. Burke, R. 2002. "Hybrid Recommender Systems: Survey and Experiments," User\nmodeling and user-adapted interaction (12:4), pp. 331-370.\n11. Cai, J.-J., Tang, J., Chen, Q.-G., Hu, Y., Wang, X., and Huang, S.-J. 2019. "MultiView Active Learning for Video Recommendation," IJCAI, pp. 2053-2059.\n12. Chen, C.-M., and Sun, Y.-C. 2012. "Assessing the Effects of Different Multimedia\nMaterials on Emotions and Learning Performance for Visual and Verbal Style\nLearners," Computers & Education (59:4), pp. 1273-1285.\n13. Chen, H.-H., Chung, C.-A., Huang, H.-C., and Tsui, W. 2017. "Common Pitfalls\nin Training and Evaluating Recommender Systems," ACM SIGKDD Explorations\nNewsletter (19:1), pp. 37-45.\n14. Choi, S.-M., Jang, K., Lee, T.-D., Khreishah, A., and Noh, W. 2020. "Alleviating\nItem-Side Cold-Start Problems in Recommender Systems Using Weak\nSupervision," IEEE Access (8), pp. 167747-167756.\n15. Collins, A., Tkaczyk, D., Aizawa, A., and Beel, J. 2018. "Position Bias in\nRecommender Systems for Digital Libraries," International Conference on\nInformation: Springer, pp. 335-344.\n16. Common-Wealth-Magazine ,2021. Complete investigation report of "2021 Listening to\nEconomic Survey" https://www.cw.com.tw/article/5115003?template=transformers (Last\naccess date 2022/07)\n77\n17. Darshna, P. 2018. "Music Recommendation Based on Content and Collaborative\nApproach & Reducing Cold Start Problem," 2018 2nd International Conference\non Inventive Systems and Control (ICISC): IEEE, pp. 1033-1037.\n18. Das, D., Sahoo, L., and Datta, S. 2017. "A Survey on Recommendation System,"\nInternational Journal of Computer Applications (160:7).\n19. Deng, S., Wang, D., Li, X., and Xu, G. 2015. "Exploring User Emotion in\nMicroblogs for Music Recommendation," Expert Systems with Applications\n(42:23), pp. 9284-9293.\n20. Díaz-Morales, J. F., Escribano, C., and Jankowski, K. S. 2015. "Chronotype and\nTime-of-Day Effects on Mood During School Day," Chronobiology international\n(32:1), pp. 37-42.\n21. Felício, C. Z., Paixão, K. V., Barcelos, C. A., and Preux, P. 2017. "A Multi-Armed\nBandit Model Selection for Cold-Start User Recommendation," Proceedings of\nthe 25th Conference on User Modeling, Adaptation and Personalization, pp. 32-\n40.\n22. Gunawan, A. A., and Suhartono, D. 2019. "Music Recommender System Based\non Genre Using Convolutional Recurrent Neural Networks," Procedia Computer\nScience (157), pp. 99-109.\n23. Guo, G., Zhang, J., and Yorke-Smith, N. 2015. "Trustsvd: Collaborative Filtering\nwith Both the Explicit and Implicit Influence of User Trust and of Item Ratings,"\nProceedings of the AAAI Conference on Artificial Intelligence.\n24. He, X., Du, X., Wang, X., Tian, F., Tang, J., and Chua, T.-S. 2018. "Outer ProductBased Neural Collaborative Filtering," arXiv preprint arXiv:1808.03912).\n25. He, X., Liao, L., Zhang, H., Nie, L., Hu, X., and Chua, T.-S. 2017. "Neural\nCollaborative Filtering," Proceedings of the 26th international conference on\nworld wide web, pp. 173-182.\n26. Herce-Zelaya, J., Porcel, C., Bernabé-Moreno, J., Tejeda-Lorente, A., and\nHerrera-Viedma, E. 2020. "New Technique to Alleviate the Cold Start Problem in\nRecommender Systems Using Information from Social Media and Random\nDecision Forests," Information Sciences (536), pp. 156-170.\n27. Herlocker, J. L., Konstan, J. A., Borchers, A., and Riedl, J. 1999. "An Algorithmic\nFramework for Performing Collaborative Filtering," Proceedings of the 22nd\nannual international ACM SIGIR conference on Research and development in\ninformation retrieval, pp. 230-237.\n28. Hidalgo, M. P., Caumo, W., Posser, M., Coccaro, S. B., Camozzato, A. L., and\nChaves, M. L. F. 2009. "Relationship between Depressive Mood and Chronotype\nin Healthy Subjects," Psychiatry and clinical neurosciences (63:3), pp. 283-290.\n29. Hu, Y., and Ogihara, M. 2011. "Nextone Player: A Music Recommendation\nSystem Based on User Behavior," ISMIR, pp. 103-108.\n30. Hyung, Z., Lee, K., and Lee, K. 2014. "Music Recommendation Using Text\nAnalysis on Song Requests to Radio Stations," Expert Systems with Applications\n(41:5), pp. 2608-2618.\n31. Inside ,2020. Show you Podcast- 2020 Taiwan Podcast Industry Report for the\nFirst Half of the Year https://www.inside.com.tw/article/20391-2020-podcastreport (Last access date 2022/07)\n32. Ismailoglu, F. 2021. "Aggregating User Preferences in Group Recommender\nSystems: A Crowdsourcing Approach," Decision Support Systems), p. 113663.\n33. Jelodar, H., Wang, Y., Rabbani, M., Ahmadi, S. B. B., Boukela, L., Zhao, R., and\nLarik, R. S. A. 2021. "A Nlp Framework Based on Meaningful Latent-Topic\n78\nDetection and Sentiment Analysis Via Fuzzy Lattice Reasoning on Youtube\nComments," Multimedia Tools and Applications (80:3), pp. 4155-4181.\n34. Joo, Y. S., and Kim, S.-K. 2020. "Scent Emotion Evaluation Experiment for\nMultimedia Application," 2020 IEEE International Conference on Consumer\nElectronics-Asia (ICCE-Asia): IEEE, pp. 1-4.\n35. Kaji, N., and Kobayashi, H. 2017. "Incremental Skip-Gram Model with Negative\nSampling," arXiv preprint arXiv:1704.03956).\n36. Kim, H.-N., Ha, I., Lee, K.-S., Jo, G.-S., and El-Saddik, A. 2011. "Collaborative\nUser Modeling for Enhanced Content Filtering in Recommender Systems,"\nDecision Support Systems (51:4), pp. 772-781.\n37. Kudielka, B. M., Schommer, N. C., Hellhammer, D. H., and Kirschbaum, C. 2004.\n"Acute Hpa Axis Responses, Heart Rate, and Mood Changes to Psychosocial\nStress (Tsst) in Humans at Different Times of Day," Psychoneuroendocrinology\n(29:8), pp. 983-992.\n38. Kumar, P., and Thakur, R. S. 2018. "Recommendation System Techniques and\nRelated Issues: A Survey," International Journal of Information Technology\n(10:4), pp. 495-501.\n39. Lang, A., and Chrzan, J. 2015. "Media Multitasking: Good, Bad, or Ugly?,"\nAnnals of the International Communication Association (39:1), pp. 99-128.\n40. Lee, J.-S., and Shin, D.-H. 2015. "A Study on the Preference between Emotion of\nHuman and Media Genre in Smart Device," Science of Emotion and Sensibility\n(18:1), pp. 59-66.\n41. Li, G., and Zhang, J. 2018. "Music Personalized Recommendation System Based\non Improved Knn Algorithm," 2018 IEEE 3rd Advanced Information Technology,\nElectronic and Automation Control Conference (IAEAC): IEEE, pp. 777-781.\n42. Li, Y.-M., Liou, J.-H., and Li, Y.-W. 2020. "A Social Recommendation Approach\nfor Reward-Based Crowdfunding Campaigns," Information & Management (57:7),\np. 103246.\n43. Liang, T.-P., Lai, H.-J., and Ku, Y.-C. 2006. "Personalized Content\nRecommendation and User Satisfaction: Theoretical Synthesis and Empirical\nFindings," Journal of Management Information Systems (23:3), pp. 45-70.\n44. Llisterri, J. 1992. "Speaking Styles in Speech Research," Workshop on Integrating\nSpeech and Natural Language: Citeseer.\n45. Marshall, D., and Sidorov, K. 2001. "Introduction to Multimedia," UK: School of\nComputer Science & Informatics Cardiff University).\n46. Martikainen, K. 2020. "Audio-Based Stylistic Characteristics of Podcasts for\nSearch and Recommendation: A User and Computational Analysis." University of\nTwente.\n47. Massquantity, 2021. LibRecommender [Source code].\nhttps://github.com/massquantity/LibRecommender. (Last access date 2022/07)\n48. Mehrabian, A., and Russell, J. A. 1974. An Approach to Environmental\nPsychology. the MIT Press.\n49. Moscato, V., Picariello, A., and Sperli, G. 2020. "An Emotional Recommender\nSystem for Music," IEEE Intelligent Systems).\n50. Murray, G., Nicholas, C. L., Kleiman, J., Dwyer, R., Carrington, M. J., Allen, N.\nB., and Trinder, J. 2009. "Nature’s Clocks and Human Mood: The Circadian\nSystem Modulates Reward Motivation," Emotion (9:5), p. 705.\n79\n51. National Audiovisual Institute, 2022.[Source code]\nhttps://github.com/ina-foss/inaSpeechSegmenter (Last access date 2022/07)\n52. Nazari, Z., Charbuillet, C., Pages, J., Laurent, M., Charrier, D., Vecchione, B., and\nCarterette, B. 2020. "Recommending Podcasts for Cold-Start Users Based on\nMusic Listening and Taste," in: Proceedings of the 43rd International ACM SIGIR\nConference on Research and Development in Information Retrieval. Virtual Event,\nChina: Association for Computing Machinery, pp. 1041–1050.\n53. Panagiotakis, C., Papadakis, H., and Fragopoulou, P. 2021. "A Dual Hybrid\nRecommender System Based on Scor and the Random Forest," Computer Science\nand Information Systems (18:1), pp. 115-128.\n54. Raheem, K. R., and Ali, I. H. 2020. "Multimodal Content-Based Recommender\nSystem Using Three-Dimension Convolution Neural Network," COMPUSOFT:\nAn International Journal of Advanced Computer Technology (9:5).\n55. Rana, M. K. C. 2012. "Survey Paper on Recommendation System,").\n56. Rendle, S. 2010. "Factorization Machines," 2010 IEEE International conference\non data mining: IEEE, pp. 995-1000.\n57. Roenneberg, T. 2012. "What Is Chronotype?," Sleep and biological rhythms (10:2),\npp. 75-76.\n58. Ruff, J. 2002. "Information Overload: Causes, Symptoms and Solutions," Harvard\nGraduate School of Education), pp. 1-13.\n59. Ryu, J., Capistrano, E. P., and Lin, H.-C. 2020. "Non-Korean Consumers’\nPreferences on Korean Popular Music: A Two-Country Study," International\nJournal of Market Research (62:2), pp. 234-252.\n60. Sanz-Cruzado, J., Castells, P., and López, E. 2019. "A Simple Multi-Armed\nNearest-Neighbor Bandit for Interactive Recommendation," Proceedings of the\n13th ACM Conference on Recommender Systems, pp. 358-362.\n61. Sarwar, B., Karypis, G., Konstan, J., and Riedl, J. 2000. "Application of\nDimensionality Reduction in Recommender System-a Case Study," Minnesota\nUniv Minneapolis Dept of Computer Science.\n62. Selvi, C., and Sivasankar, E. 2019. "An Efficient Context-Aware Music\nRecommendation Based on Emotion and Time Context," in Data Science and Big\nData Analytics. Springer, pp. 215-228.\n63. Shani, G., and Gunawardana, A. 2011. "Evaluating Recommendation Systems,"\nin Recommender Systems Handbook. Springer, pp. 257-297.\n64. Son, J., and Kim, S. B. 2018. "Academic Paper Recommender System Using\nMultilevel Simultaneous Citation Networks," Decision Support Systems (105), pp.\n24-33.\n65. Subramaniyaswamy, V., and Logesh, R. 2017. "Adaptive Knn Based\nRecommender System through Mining of User Preferences," Wireless Personal\nCommunications (97:2), pp. 2229-2247.\n66. Sulaiman, N., Muhammad, A. M., Ganapathy, N. N. D. F., Khairuddin, Z., and\nOthman, S. 2017. "A Comparison of Students` Performances Using Audio Only\nand Video Media Methods," English language teaching (10:7), pp. 210-215.\n67. Taylor, B. J., and Hasler, B. P. 2018. "Chronotype and Mental Health: Recent\nAdvances," Current psychiatry reports (20:8), pp. 1-10.\n68. Thales,2021. PySpeech [Source code]\nhttps://github.com/thalesaguiar21/PySpeech (Last access date 2022/07)\n69. Thorat, P. B., Goudar, R., and Barve, S. 2015. "Survey on Collaborative Filtering,\nContent-Based Filtering and Hybrid Recommendation System," International\nJournal of Computer Applications (110:4), pp. 31-36.\n80\n70. Tsagkias, M., Larson, M., and De Rijke, M. 2010. "Predicting Podcast Preference:\nAn Analysis Framework and Its Application," Journal of the American Society for\ninformation Science and Technology (61:2), pp. 374-391.\n71. Volkovs, M., Yu, G. W., and Poutanen, T. 2017. "Content-Based Neighbor\nModels for Cold Start in Recommender Systems," in Proceedings of the\nRecommender Systems Challenge 2017. pp. 1-6.\n72. Volokhin, S., and Agichtein, E. 2018. "Understanding Music Listening Intents\nDuring Daily Activities with Implications for Contextual Music\nRecommendation," Proceedings of the 2018 Conference on Human Information\nInteraction & Retrieval, pp. 313-316.\n73. Wang, D., Zhang, X., Yu, D., Xu, G., and Deng, S. 2020a. "Came: Content-and\nContext-Aware Music Embedding for Recommendation," IEEE transactions on\nneural networks and learning systems (32:3), pp. 1375-1388.\n74. Wang, R., Ma, X., Jiang, C., Ye, Y., and Zhang, Y. 2020b. "Heterogeneous\nInformation Network-Based Music Recommendation System in Mobile\nNetworks," Computer Communications (150), pp. 429-437.\n75. Wang, X., Rosenblum, D., and Wang, Y. 2012. "Context-Aware Mobile Music\nRecommendation for Daily Activities," Proceedings of the 20th ACM\ninternational conference on Multimedia, pp. 99-108.\n76. Xing, Z., Parandehgheibi, M., Xiao, F., Kulkarni, N., and Pouliot, C. 2016.\n"Content-Based Recommendation for Podcast Audio-Items Using Natural\nLanguage Processing Techniques," 2016 IEEE International Conference on Big\nData (Big Data): IEEE, pp. 2378-2383.\n77. Xu, L., Wen, X., Shi, J., Li, S., Xiao, Y., Wan, Q., and Qian, X. 2021. "Effects of\nIndividual Factors on Perceived Emotion and Felt Emotion of Music: Based on\nMachine Learning Methods," Psychology of Music (49:5), pp. 1069-1087.\n78. Xu, X., Dutta, K., and Ge, C. 2018. "Do Adjective Features from User Reviews\nAddress Sparsity and Transparency in Recommender Systems?," Electronic\nCommerce Research and Applications (29), pp. 113-123.\n79. Yang, L., Sobolev, M., Wang, Y., Chen, J., Dunne, D., Tsangouri, C., Dell, N.,\nNaaman, M., and Estrin, D. 2019. "How Intention Informed Recommendations\nModulate Choices: A Field Study of Spoken Word Content," The World Wide Web\nConference, pp. 2169-2180.\n80. Ye, B. K., Tu, Y. J. T., and Liang, T. P. 2019. "A Hybrid System for Personalized\nContent Recommendation," Journal of Electronic Commerce Research (20:2), pp.\n91-104.\n81. Yin, H., Cui, B., Li, J., Yao, J., and Chen, C. 2012. "Challenging the Long Tail\nRecommendation," arXiv preprint arXiv:1205.6700).\n82. Yu, T., Guo, J., Li, W., and Lu, M. 2021. "A Mixed Heterogeneous Factorization\nModel for Non-Overlapping Cross-Domain Recommendation," Decision Support\nSystems), p. 113625.\n83. Zhang, H., Lu, Y., Gupta, S., and Zhao, L. 2014. "What Motivates Customers to\nParticipate in Social Commerce? The Impact of Technological Environments and\nVirtual Customer Experiences," Information & Management (51:8), pp. 1017-\n1030.\n84. Zhu, Y., Lin, J., He, S., Wang, B., Guan, Z., Liu, H., and Cai, D. 2019. "Addressing\nthe Item Cold-Start Problem by Attribute-Driven Active Learning," IEEE\nTransactions on Knowledge and Data Engineering (32:4), pp. 631-644 zh_TW dc.identifier.doi (DOI) 10.6814/NCCU202201103 en_US
