(18.204.2.190) 您好!臺灣時間:2021/04/19 07:50
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:劉雅茹
研究生(外文):Ya-Ju Liu
論文名稱:利用本體論建構群組偏好樹於產品推薦之研究
論文名稱(外文):Using Ontology and Group Preference Tree for Product Recommendation
指導教授:李麗華李麗華引用關係
指導教授(外文):Li-Hua Li
學位類別:碩士
校院名稱:朝陽科技大學
系所名稱:資訊管理系碩士班
學門:電算機學門
學類:電算機一般學類
論文種類:學術論文
論文出版年:2007
畢業學年度:95
語文別:中文
論文頁數:97
中文關鍵詞:正規化概念分析自適應共振理論類神經網路本體論推薦系統產品推薦
外文關鍵詞:Recommendation SystemAdaptive Resonance Theory (ART)Formal Concept Analysis (FCA)Product RecommendationArtificial Neural Network (ANN)Ontology
相關次數:
  • 被引用被引用:7
  • 點閱點閱:641
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:126
  • 收藏至我的研究室書目清單書目收藏:3
在推薦系統的領域中,學者們提出許多方法與技術,其中有利用資料探勘(Data Mining)技術分析顧客的購買行為或是利用類神經網路(Artificial Neural Network, ANN)方法預測顧客的喜好, 連近來很熱門的本體論(Ontology)技術也漸漸地被運用在推薦系統(Recommendation System)上,這是因為本體論它的階層間有明確的定義與關係,除了能有效的挖掘顧客潛在偏好外,透過概念的清楚定義能取代模糊不清的關鍵字,有效且準確的預測顧客喜愛的產品,藉以提昇推薦品質。
在過去研究中Middleton (2004)等學者提出一套以本體論技術為基礎的論文推薦系統Quickstep[40],採用本體論技術建構使用者的個人輪廓檔(User Profile),將使用者對於論文的喜好程度記錄於個人輪廓檔中。除此之外,尚有運用本體論及個人化技術於網際網路環境的通用個人輪廓(Universal User Profile, UUP),以供使用者遨遊於不同網站時,能便利的取得個人化的網路服務[9],然而這些研究皆是利用本體論技術來建構個人偏好,較少談論到群組偏好的議題,記錄使用者個人偏好固然可以達到個人化之目的,但必須在使用者個人資料充裕的情況下方能加以運用。
有鑑於此,本研究以提出一個結合類神經網路與本體論的推薦機制,以產品銷售資料為推薦之實驗,做為檢驗本論文所提之機制能提高產品推薦的準確率。本研究提出的方法為運用自適應共振理論網路(Adaptive Resonance Theory, ART)之聚類特性,先建立群組偏好樹(Group Preference Tree, GPT),透過此偏好樹,推論顧客對於產品的喜好強度,利用事先聚類的方法,將顧客依其屬性分群並協同推薦,以提高推薦效能。本研究所提之方法具有下列好處:(1) 透過群組偏好樹之建立,即使顧客從未購買過某類產品,也可以運用本體推論方法(Ontology Inference),分享同群成員的偏好,加以適當的推薦,期能增加顧客購買量與推薦效能;(2) 改善現有本體論推薦系統(Ontological Recommendation System)的不足,加入分群之概念,因此即使個人資料量稀少,也能透過分群後所建立的群組偏好樹與同群成員之間的偏好分享,得到推薦之效果,降低資料稀疏性(Sparsity)問題。為驗證本研究之實務性與可行性,本研究將利用食品超市之產品交易資料來驗證本研究的準確度與降低資料的稀疏性之效果,最後,由本研究的實驗可看出,利用群組偏好樹及本研究所提之方法步驟,確能有效推薦商品,挖掘使用者潛在需求,改善資料稀疏性所帶來的問題。
In the recent years, various techniques of implementing recommendation systems have been proposed for analyzing information. These techniques, which include Artificial Neural Network (ANN) and Data Mining (DM), are applied to analyze user’s purchasing behaviors and to predict user’s interests. To further enhance the information classification task, the recent study has shown that ontology can be an emerging technology for recommendation by its explicit specification of conceptualization. Ontology can be an effective method for data representation. Not only the user’s potential demand can be discovered, but also the ambiguous keywords can be replaced. Therefore, the user’s interests can be efficiently predicted to improve the recommendation quality.
In the past research, Middleton et al. [40] have proposed a novel ontological approach to construct user profile within the recommender system. This system was dealing with the recommending of academic papers on-line and this system is called Quickstep. The user profile also recorded user’s interest. In addition to Quickstep, Ren-De Ou [9] also proposed an approach called Universal User Profile (UUP), which used ontology and personalization approach to perform the recommendation. However, this research focused only on individual preference using ontology technique. And the group preference was not considered. As a result, the group preferences are not utilized especially when using the collaborating filtering technique for recommendation. This is certainly a miss for not incorporating the group ontology for discovering the purchasing behavior.
To adopt the ontology method in the recommendation system for better outcome, this research proposes the PROPT (Product Recommendation using Ontology Preference Tree) system. The PROPT system utilizes ANN to reveal the features of users by clustering user profiles, building the product ontology to construct the Group Preference Tree (GPT), and, finally, applying ontology inference to make recommendation.
To demonstrate and to show the performance of PROPT system, this research has conducted an experiment with 2000 users and 110 products. The experiment result has shown that this research can achieve the followings,
(1) to recommend products with better efficiency,
(2) to resolve the problem of data sparsity,
(3) to dig out the potential demand of users.
摘要 I
Abstract III
誌謝 V
目錄 VII
表目錄 X
圖目錄 XI
第一章 緒論 1
1.1 導論 1
1.2 研究目的 3
1.3 研究範圍 4
1.4 論文架構 5
第二章 文獻探討 7
2.1 推薦系統(Recommendation System) 7
2.1.1 內容導向式(Content Based, CB)推薦 7
2.1.2 協同過濾(Collaborative Filtering, CF)推薦 8
2.1.3 混合式推薦(Hybrid Approach Recommendation) 10
2.2 類神經網路(Artificial Neural Network, ANN) 11
2.2.1 類神經網路的網路型態 11
2.2.2 自適應共振理論(Adaptive Resonance Theory, ART) 12
2.2.2.1 ART學習演算法 14
2.2.2.2 膝點(Steepest Knee)判斷法 18
2.3 個人輪廓(User Profile, UP) 18
2.3.1 定義(Definition) 19
2.3.2 個人輪廓相關研究 20
2.4 本體論(Ontology) 23
2.4.1 定義(Definition) 23
2.4.2 本體論概念推論(Ontology Inference) 24
2.4.3 本體論推薦系統相關研究 27
2.5 正規化概念分析(Formal Concept Analysis, FCA) 29
2.5.1 定義(Definition) 29
2.5.2 概念矩陣(Concept Lattice) 30
2.5.3 FCA-based相似度測量(FCA-based Similarity Measure) 32
2.6 推薦效能評估 36
第三章 研究方法 37
3.1 PROPT系統架構 37
3.2 研究方法與步驟 39
第四章 實驗與分析 58
4.1 實驗環境與使用工具 58
4.2 實驗資料來源 59
4.3 資料研析 59
4.4 實驗步驟 62
4.4.1使用者的個人輪廓建立 62
4.4.2 編碼與分群 63
4.4.2.1 編碼轉換 63
4.4.2.2 警戒值設定 65
4.4.2.3 分群結果 66
4.4.3 建構群組偏好樹 67
4.4.4 推論結果 68
4.5 推薦效能評估 70
4.5.1偏好推論評估之比較 70
4.5.2購買次數與購買數量之評估比較 74
第五章 結論與未來研究 76
5.1 研究貢獻 77
5.2 未來研究 78
參考文獻 80
附錄一 、產品本體論(Product Ontology) 87
附錄二 、分群結果 91

表目錄
表一、ART網路的相關變數說明 15
表二、個人輪廓相關研究說明 22
表三、本體論推薦系統相關研究說明 28
表四 、FCA定義 29
表五、個人輪廓屬性資料範例說明 40
表六、屬性轉換位元對照表 41
表七、使用工具與其任務說明 59
表八、顧客基本資料表(僅列舉部份) 60
表九、顧客交易資料表(列舉) 61
表十、六個屬性的二元編碼資料對照表 64
表十一 、分群結果(僅列舉6群為例) 66
表十二、各類別的產品權重值 67
表十三、推論結果 69

圖目錄
圖一、自適應共振理論網路架構 13
圖二、分群評估值的膝點 18
圖三、ANATAGONOMY細部架構圖 20
圖四、產品本體論簡例 24
圖五、Quickstep推論公式 26
圖六、網頁推薦推論公式 27
圖七、Context (O: User A: Movie) 31
圖八、Concept Lattice (O: User A: Movie) 31
圖九、Attribute-based相似度測量公式 34
圖十、FCA概念相似度測量公式 35
圖十一、PROPT系統架構圖 37
圖十二、PROPT系統之處理流程 38
圖十三、個人輪廓建立流程 39
圖十四、編碼轉換流程 40
圖十五、運用ART作分群之流程 42
圖十六、建構GPT流程圖 45
圖十七、Product Ontology Tree 46
圖十八、群組偏好樹之購買權重計算範例 47
圖十九、由群組偏好樹找出最大權重之節點 47
圖二十、GPT推論流程圖 48
圖二十一、偏好內文(Preference Context)-簡例 51
圖二十二、概念矩陣-FCA-based GPT 52
圖二十三、第t群的群組偏好樹(GPTt) 57
圖二十四、第t群的FCA-based GPT 57
圖二十五、實驗步驟流程圖 61
圖二十六、警戒值-群數變動情況 65
圖二十七、第15群群組偏好樹(GPT15) 69
圖二十八、Group、Personal與Common Preference之回應率比較圖71
圖二十九、Group、Personal與Common Preference之準確率比較圖72
圖三十、Group、Personal與Common Preference之F1指標比較圖73
圖三十一、Group、Personal與Common Preference之人數比較圖74
圖三十二 、購買次數與購買數量之評估比較圖 75
[1]王奕涵(2006),正規化概念分析的資訊管理領域理論之知識本體建構,碩士論文,國立雲林科技大學資訊管理系,雲林。
[2]李季錦(2003),導引國中教師線上編輯教材之研究──以CRMS為基礎,碩士論文,國立高雄師範大學資訊教育研究所,高雄。
[3]林信成(2000),「軟性計算理論與知識處理技術之研究」,教育資料與圖書館學,第三十八卷,第二期,第148-173頁。
[4]邱永祥(2003),運用類神經網路與資料探勘技術於網路教學課程推薦之研究,碩士論文,朝陽科技大學資訊管理學系,臺中。
[5]許昌齡(2001),「資料分類、群聚及其整合之研究:從資料挖掘觀點」,九十年度弘光技術學院學術研究發表會論文集,臺中,第136-151頁。
[6]郭盛揚(2003),以顧客交易特徵為基礎的顧客分群方法,碩士論文,天主教輔仁大學資訊管理學系,臺北。
[7]黃鈁媖(2006),運用類神經網路預測新進顧客產品喜好之個人化商品推薦技術,碩士論文,朝陽科技大學資訊管理學系,臺中。
[8]葉怡成(2003),類神經網路模式應用與實作,儒林圖書有限公司,第9-1 - 9-29頁。
[9]歐仁德(2006),結合本體論與通用個人輪廓於個人化推薦之研究,碩士論文,朝陽科技大學資訊管理學系,臺中。
[10]鄭丞君(2004),資料挖掘技術在旅遊行銷之應用,碩士論文,南華大學資訊管理學系,嘉義。
[11]謝昆霖、謝明峰(2005),「具智慧型媒合機制之人力資源網站建置與應用」,中華管理評論國際學報,第八卷,第四期,第1-28頁。
[12]N. B. Abdallah (1993), “Description de Documents Textuels: Indices Pour Une Typologies Present en Compte le Contexte et la Finalite de la Communication,” Memorie de DEA Sciennce de l''information et de la commmunication, University Lyon II, ENSSIB.
[13]W. N. Borst (1997), “Construction of Engineering Ontologies for Knowledge Sharing and Reuse,” Ph.D. Dissertation, University of Twente, Enschede, The Netherlands.
[14]G. A. Carpenter and S. Grossberg (1998), The Handbook of Brain Theory and Neural Networks. Adaptive Resonance Theory, In M.A. Arbib (ed), Second Edition, Cambridge, MA: MIT Press, ISBN: 0-262-51102-9, pp. 79-82.
[15]M. S. Chen, J. Han, and P. S. Yu (1996), “Data Mining: An Overview from a Database Perspective,” IEEE Trans. on Knowledge and Data Engineering, Vol. 8, No. 6, pp. 866-883.
[16]F. Chesani (2002), “Recommendation Systems,” Corso di laurea in Ingegneria Informatica, pp. 1-32.
[17]Y. B. Cho, Y. H. Cho, and S. H. Kim (2005), “Mining Changes in Customer Buying Behavior for Collaborative Recommendations,” Expert Systems with Applications, Vol. 28, No. 4, pp. 359-369.
[18]P. J. Daniels (1986), “Cognitive Models in Information Retrieval-An Evaluative Review,” Journal of Documentation, Vol. 42, No. 4, pp. 272-304.
[19]F. Dau and J. Klinger (2005), “From Formal concept analysis to contextual logic,” In Proceedings of International Conference on Formal Concept Analysis (ICFCA).
[20]P. du Boucher-Ryan and D. Bridge (2006), “Collaborative Recommending Using Formal Concept Analysis,” Knowledge-Based Systems, Vol. 19, pp. 309-315.
[21]G. S. Francisco, M. B. Rodrigo, and C. Leonardo (2006), “An Ontology-based Intelligent System for Recruitment,” Expert Systems with Applications, Vol. 31, pp. 248-263.
[22]A. Formica (2006), “Ontology-based Concept Similarity in Formal Concept Analysis,” Information Sciences, Vol. 176, pp. 2624-2641.
[23]A. Formica and M. Missikoff (2002), “Concept Similarity in SymOntos: An Enterprise Ontology Management Tool,” The Computer Journal, Vol. 45, No. 6, pp. 583-594.
[24]B. Ganter and R. Wille (1999), Formal concept anlysis - mathematical foundations, Verlag: Springer.
[25]D. Goldberg, D. Nichols, B. M. Oki, and D. Terry (1992), “Using Collaborative Filtering to Weave an Information TAPESTRY,” Communications of the ACM, Vol. 35, No. 12, pp. 61-70.
[26]S. Grossberg (1976), “Adaptive Pattern Classification and Universal Recoding: I. Parallel Development and Coding of Neural Feature Detectors,” Biological Cybernetics, Vol. 23, pp. 121-134.
[27]T. R. Gruber (1993), “A Translation Approach to Portable Ontology Specifications,” Knowledge Acquisition, Vol. 5, pp. 199-220.
[28]J. Han and M. Kamber (2000), Data Mining: Concepts and Techniques, Morgan Kaufmann Publishers.
[29]J. J. Hopfield (1982), “Neural Networks and Physical Systems with Emergent Collective Computational Abilities,” Proceedings of the National Academy of Sciences of the United States of America, Vol. 79, pp. 2554-2558.
[30]L. P. Hung (2005), “A Personalized Recommendation System based on Product Taxonomy for One-to-One Marketing Online,” Expert Systems with Applications, Vol. 29, No. 2, pp. 383-392.
[31]E. Hyvönen, E. Mäkelä, M. Salminen, A. Valo, K. Viljanen, S. Saarela, M. Junnila, and S. Kettula (2005), “MUSEUMFINLAN─Finnish Museums on the Semantic Web,” Journal of Web Semantics, Vol. 3, pp. 224-241.
[32]G. Jiang, K. Ogasawara, A. Endoh, and T. Sakurai (2003), “Context-based Ontology Building Support in Clinical Domains Using Formal Concept Analysis,” International Journal of Medical Informatics, Vol. 71, No.1, pp. 71-81.
[33]Y. S. Kim, B. J. Yum, J. Song, and S. M. Kim (2005), “Development of a Recommender System Based on Navigational and Behavioral Patterns of Customers in E-Commerce Sites,” Expert Systems with Applications, Vol. 28, pp. 381-393.
[34]T. Kohonen (1989), Self-Organization and Associative Memory, 3rd edition, Springer-Verlag.
[35]O. B. Kwon (2003), “I Know What You Need to Buy: Context-Aware Multimedia-based Recommendation System,” Expert Systems with Applications, Vol. 25, pp. 387-400.
[36]S. Laine-Cruzel, T. Lafouge, J. P. Lardy, and N. B. Abdallah (1996), “Improving Information Retrieval by Combining User Profile and Document Segmentation,” Information Processing & Management, Vol. 32, No. 3, pp. 305-315.
[37]Y. Li, L. Lu, and L. Xuefeng (2005), “A Hybrid Collaborative Filtering Method for Multiple-Interests and Multiple-Content Recommendation in E-Commerce,” Expert Systems with Applications, Vol. 28, No. 1, pp. 66-77.
[38]W. S. McCulloch and E. Pitts (1943), “A Logical Calculus of the Ideas Imminent in Nervous Activity,” Bull Math Biophy, Vol. 5, pp. 115-133.
[39]P. Melville and R. Nagarajan (2001), “Content-Boosted Collaborative Filtering,” Proceedings of the ACM SIGIR-2001 Workshop on Recommender Systems, New Orleans, LA, September, pp. 1-9.
[40]S. E. Middleton, N. R. Shadbolt, and D. C. De Roure (2004), “Ontological User Profiling in Recommender Systems,” ACM Transactions on Information Systems (TOIS), Vol. 22, No. 1, pp. 54-88.
[41]R. J. Mooney and L. Roy (1999), “Content-based Book Recommending Using Learning for Text Categorization,” Proceedings of the SIGIR-99 Workshop on Recommender Systems: Algorithms and Evaluation, Bekeley CA, pp. 195-204.
[42]N. F. Noy and D. L. McGuinness (2001), “Ontology Development 101: A Guide to Creating Your First Ontology,” Stanford Knowledge Systems Laboratory Technical Report, KSL-01-05, and Stanford Medical Informatics Technical Report, SMI-2001-0880.
[43]M. Papagelis and D. Plexousakis (2005), “Qualitative Analysis of User-based and Item-based Prediction Algorithms for Recommendation Agents,” Engineering Applications of Artificial Intelligence, Vol. 18, No. 4, pp. 781-789.
[44]M. J. Pazzni (1999), “A Framework for Collaborative, Content-based, and Demographic Filtering,” Artificial Intelligence Review, Vol. 13, No. 5-6, pp. 393-408.
[45]P. Resnick, N. Iacovou, M. Suchak, P. Bergstrom, and J. Riedl (1994), “Grouplens: An Open Architecture for Collaborative Filtering of Netnews,” Proceedings of the ACM 1994 Conference on Computer Supported Cooperative Work, pp. 175-186.
[46]C. J. Rijsbergen (1979), Information Retrieval, 2nd ed., London: Butterworths.
[47]H. Sakagami and T. Kamba (1997), “Learning Personal Preferences on Online Newspaper Articles from User Behaviors,” Computer Networks and ISDN Systems, Vol. 29, pp. 1447-1455.
[48]J. B. Schafer, J. Konstan, and J. Riedl (1999), “Recommender Systems in E-Commerce,” Proceedings of the First ACM Conference on Electronic Commerce, pp. 158-166.
[49]U. Shardanand and P. Maes (1995), “Social Information Filtering: Algorithms for Automating “Word of Mouth”,” Proceedings of the Conference on Human Factors in Computing Systems, pp. 210-217.
[50]N. Weiβenberg, A. Voisard, and R. Gartmann (2004), “Using Ontologies in Personalized Mobile Applications,” Proceedings of the International ACM GIS Conference (GIS’04), Washington D.C., USA, pp. 2-11.
[51]R. Wille (1982), Restructuring Lattice Theory: An Approach based on Hierarchies of Concepts, in: I. Rival (Ed.), Symposium on Ordered Sets, Reidel, Dordrecht, Boston.
[52]S. T. Yuan and C. Cheng (2004), “Ontology-based Personalized Couple Clustering for Heterogeneous Product Recommendation in Mobile Marketing,” Expert Systems with Applications, Vol.26, pp. 461-476.
[53]Y. Zhao and W. Halang (2006), “Rough Concept Lattice based Ontology Similarity Measure,” Proceedings of the First International Conference on Scalable Information Systems (INFOSCALE’06), Hong Kong, pp. 1-7.
[54]Amazon.com (2006), http://www.amazon.com.
[55]CDNow.com (2006), http://www.cdnow.com.
[56]eBay (2006), http://www.eBay.com.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔