臺灣博碩士論文加值系統

推薦系統目的在於收集大眾對於資源的評價，經過推薦機制的整理，呈現給符合偏好的使用者，但是Supernode架構與傳統主從架構的不同，增加推薦系統收集、統計推薦指標的困難。在本篇論文中，將利用RDF(S)建置與Supernode、Peer、MP3相關的本體論，同時，依據Supernode與Peer在推薦系統中所扮演的功能與角色的不同，將不同的結構藍圖分別放置在Supernode與Peer之中，並且藉由Supernode與Peer中所描述的資源、屬性以及資源和資源之間的關係來解決在Supernode架構中推薦指標的收集、統計與更新等問題，以及降低推薦系統中"Cold Start"與"Spamming"這兩個問題所產生的影響。

The purpose of recommendation system is to collect public opinions and provide synthetic recommended result for users. Using our proposed recommended mechanism, the recommendation system presents personal music recommended list for each user in peer node to match his preference. Because the complexity difference of network structure, we found that the rating collection and computation process will be more complex in supernode P2P network than in client-server WWW network. In this thesis, firstly we use RDF(S) to construct associated ontology schema for supernode, peer node, and MP3 music. Then, we put respective ontology schema in the suprenode and peer node based on their roles and functions within the recommendation system. The resources and attributes relationships between supernode and peer nodes provide the capacity to solve the rating collection, analysis, and updating problems. We also propose possible solutions for reducing the side effects of cold start and spamming problems in the recommendation system.

第一章 導論 1
1.1研究背景 1
1.2研究目的 4
1.3已完成項目 5
1.4各章節闡述 5
第二章 相關研究 6
2.1推薦系統 6
2.2推薦系統與對等式網路 7
2.3語意網與推薦系統 8
2.4語意網與對等式網路 8
2.5語意網結合對等式網路的應用系統 9
第三章 推薦系統本體論設計 11
3.1 主從架構下的推薦系統 11
3.2 Supernode架構中的推薦系統 12
3.3推薦系統本體論 16
3.3.1本體論設計 16
3.3.2本體論存放 21
第四章 推薦機制的設計 24
4.1 Peer歸屬曲風的界定 24
4.2推薦指標的收集 25
4.3推薦指標的設計 26
4.3.1 Peer評分比重計算 26
4.3.2時間遞移變數 27
4.3.3推薦指標的計算 28
4.3.4 推薦指標的統計時機 31
4.3.5 Cold Start Problem 32
4.3.6 Spamming Problem 33
4.4推薦清單的個人化 34
4.5 MP3動向的掌握 35
第五章 系統架構 36
5.1系統架構 36
5.2 推薦系統流程 37
5.3 Jena 38
5.4 SPARQL語法 40
5.4.1 SPARQL Use Case-獲得推薦清單 40
5.4.2 SPARQL Use Case-查詢特定歌曲 42
5.4.3 SPARQL Use Case-給予推薦指標 43
第六章 系統實作與分析 47
6.1實驗場景 47
6.2實驗系統正確性 48
6.2.1推薦清單的正確性 48
6.2.2查詢歌曲的正確性 50
6.2.3給予推薦指標的正確性 52
6.3定量分析 54
第七章 結論與未來展望 61
7.1結論 61
7.2未來展望 62
參考文獻 64
附錄一 67
附錄二 67
附錄三 68

【1】 ARQ, http://jena.sourceforge.net/ARQ/
【2】 Avesani, P. and P. Massa. (2002). Recommendation and Personalization From Item to Collection, from Client-Server to Peer-to-peer New Frontiers for Case-based Personalization. 6th European Conference on Case Based Reasoning ECCBR.
【3】 Broekstra, J., et al., A Metadata Model for Semantics-Based Peer-to-Peer Systems, http://www.cs.vu.nl/~frankh/postscript/P2P@WWW03.pdf.
【4】 Cai, M., and M. Frank. ( 2004). RDFPeers: A Scalable Distributed RDF Repository based on A Structured PeertoPeer Network, WWW 2004.
【5】 Chen, H. C. and A. L. P. Chen. (2001). A music recommendation system based on music data grouping and user interests. the tenth international conference on Information and knowledge management.
【6】 Fang-Fei Kuo and Man-Kwan Shan. (2002). A Personalized Music Filtering System Based on Melody Style Classification. IEEE International Conference on Data Mining (ICDM’02).
【7】 Georgios Pitsilis and Lindsay Marshall. (2005). Trust as a key to improving Recommendation System. Trust Management: Third International Conference, iTrust 2005.
【8】 Jena, http://jena.sourceforge.net/
【9】 Khan, L. and D. McLeod. (2000) Audio Structuring and Personalized Retrieval Using Ontologies. IEEE Advances in Digital Libraries, Library of Congress.
【10】 Mario, S., et al. (2002). HyperCuP - Hypercubes, Ontologies and Efficient Search on Peer-to-peer Networks. http://www.kbs.uni-hannover.de/Arbeiten/Publikationen/2002/hypercup-ap2pc02.pdf.
【11】 Music Recommendation System for iTunes, http://music.cs.uiuc.edu/
【12】 Nejdl, W. , et al. (2002). Edutella: A P2P Networking Infrastructure Based on RDF. WWW 2002.
【13】 Nejdl, W., et al. (2003). Super-Peer -Based Routing and Clustering Strategies for RDF-Based Peer-to-Peer Networks. Web Semantics 2003.
【14】 Parkhomenko, O., Y. Lee, and E. K. Park. (2003). Ontology-Driven Peer Profiling in Peer-to-Peer Enabled Semantic Web. CIKM’03.
【15】 Peter Haase1, et al. (2004). Bibster – A Semantics-Based Bibliographic Peer-to-Peer System. ISWC 2004.
【16】 Planet-Lab, http://www.planet-lab.org/
【17】 Pouwelse, J., et al. (2005). P2Pbased PVR Recommendation using Friends, Taste Buddies and Superpeers. IUI’05.
【18】 Pretschne, A., and S. Gauch. (1999). Ontology Based Personalized Search. 11th IEEE Intl. Conf. on Tools with Artificial Intelligence.
【19】 Qing Li,Byeong Man Kim. (2003). Clustering Approach for Hybrid Recommender System. the IEEE/WIC International Conference on Web Intelligence (WI’03).
【20】 Robin B. and et al. (2005). Identifying Attack Models for Secure Recommendation. IUI'05.
【21】 Stuart E. Middleton, and et al. (2002). Exploiting Synergy Between Ontologies and Recommender Systems. Semantic Web Workshop 2002.
【22】 SWAP (Semantic Web and P2P), http://swap.semanticweb.org/.