在協同過濾系統(Collaborative Filtering)中，我們觀察過去的研究，大部分都是以系統所收集資料(Local Resources)來做為分析基礎，採用使用者評分矩陣(Rating Matrix)來做相似性的分析和預測。像是以項目為基礎(Item-Based)協同過濾的效能和正確性，得完全依靠評分矩陣(Rating Matrix)的資料收集量及完整性而決定，當資料量不足時，就會遇到稀疏性問題(Sparsity Problem)的問題，而冷啟動(Cold-Start)則是以系統所收集資料(Local Resources)為基礎的分析條件下所無法避免的問題。
本篇論文提出了一個新的觀點，我們希望能夠找到一個額外的資料庫，來輔助以項目為基礎(Item-Based)協同過濾，不論是一般情況下，又或者是當遇到稀疏矩陣和新商品加入時，能夠使用這個額外的資料庫來計算出更準確的相似度，並結合兩個不同資料基礎的預測結果，以增加最後預測或推薦成功的準確性。
我們利用全球資訊網(www)這個的現成的龐大資料庫來當作外部資料(Global Resources)來源，在網際網路中眾多的評論、討論等資訊，越常被放在同一篇文章所討論或評判的兩商品，代表兩者之間擁有越高的相似度，本論文利用Google Similarity的計算，來求得在www中所反映的兩商品之間的相似度資訊，減緩只使用現有資料(Local Resources)所產生的問題。

Based on the previous research, mostly we applied the Local Resources as the basic analysis in Collaborative Filtering, adopting the Rating Matrix for the analysis and prediction of similarities. For example, the efficacy and the correctness of Item-Based is exclusively determined by the quantity of the collected data and the completeness of Rating Matrix. When the quantity is insufficient, it might cause the Sparsity Problem, and the Cold-Start is another inevitable problem caused by the analysis of Local Resources.
We argued for a new perspective that finding an extra database to assist the Item-Based Collaborative Filtering. No matter under which circumstances, the normal one or encountering the arsematrix and new product, we could apply the extra database to calculate the similarity more accurately, combining the predictions of the two different database to increase the accuracy and success of the final prediction.
We utilize the existed huge database, www, as Global Resources. Within the numerous comment and discussion on the Internet, the more frequently compared or discussed between the two products, the higher similarities they have. In the previous study, with the calculation of the Google Similarity, we gained the similarity information between the two products reflected in www, to soften the problem of adopting Local Resources alone.

[1] M. Balabanović and Y. Shoham, “Fab: content-based, collaborative recommendation,” Commun. ACM, vol. 40, no. 3, pp. 66–72, 1997.
[2] J. L. Herlocker, J. A. Konstan, A. Borchers, and J. Riedl, “An Algorithmic Framework for Performing Collaborative Filtering,” in Proceedings of the 22Nd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 1999, pp. 230–237.
[3] M. Papagelis and D. Plexousakis, “Qualitative analysis of user-based and item-based prediction algorithms for recommendation agents,” Eng. Appl. Artif. Intell., vol. 18, no. 7, pp. 781–789, 2005.
[4] J. Wang, A. P. de Vries, and M. J. T. Reinders, “Unifying user-based and item-based collaborative filtering approaches by similarity fusion,” Proc. 29th Annu. Int. ACM SIGIR Conf. Res. Dev. Inf. Retr. - SIGIR ’06, p. 501, 2006.
[5] J. a. Konstan, B. N. Miller, D. Maltz, J. L. Herlocker, L. R. Gordon, and J. Riedl, “GroupLens: applying collaborative filtering to Usenet news,” Commun. ACM, vol. 40, no. 3, pp. 77–87, 1997.
[6] B. Sarwar, G. Karypis, J. Konstan, and J. Riedl, “Item-based Collaborative Filtering Recommendation Algorithms,” in Proceedings of the 10th International Conference on World Wide Web, 2001, pp. 285–295.
[7] G. Linden, B. Smith, and J. York, “Amazon.com recommendations: Item-to-item collaborative filtering,” IEEE Internet Comput., vol. 7, no. 1, pp. 76–80, 2003.
[8] R. Cilibrasi and P. M. B. Vitanyi, “Automatic Meaning Discovery Using Google,” in Kolmogorov Complexity and Applications, 2006, no. 06051.
[9] R. L. Cilibrasi and P. M. B. Vitányi, “The Google similarity distance,” IEEE Trans. Knowl. Data Eng., vol. 19, no. 3, pp. 370–383, 2007.
[10] P. Resnick, H. R. Varian, and G. Editors, “Recommender Systems mmende tems,” Commun. ACM, vol. 40, no. 3, pp. 56–58, 1997.
[11] K. Lang, “NewsWeeder: Learning to Filter Netnews,” in ICML, 1995, vol. 5000, pp. 331–339.
[12] R. Armstrong, D. Freitag, and T. Joachims, “Webwatcher: A learning apprentice for the world wide web,” in AAAI Spring symposium on Information gathering from Heterogeneous, distributed environments, 1995, pp. 6–12.
[13] B. Krulwich and C. Burkey, “The InfoFinder agent: Learning user interests through heuristic phrase extraction,” IEEE Expert. Syst. their Appl., vol. 12, no. 5, pp. 22–27, 1997.
[14] R. D. Lawrence, G. S. Almasi, V. Kotlyar, M. S. Viveros, and S. S. Duri, “Personalization of Supermarket Product Recommendations,” Data Min. Knowl. Discov., vol. 5, pp. 11–32, 2001.
[15] G. Adomavicius and A. Tuzhilin, “Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions,” IEEE Transactions on Knowledge and Data Engineering, vol. 17, no. 6. pp. 734–749, 2005.
[16] D. Goldberg, D. Nichols, B. M. Oki, and D. Terry, “Using collaborative filtering to weave an information tapestry,” Communications of the ACM, vol. 35, no. 12. pp. 61–70, 1992.
[17] P. Resnick, N. Iacovou, M. Suchak, P. Bergstrom, and J. Riedl, “GroupLens : An Open Architecture for Collaborative Filtering of Netnews,” in Proceedings of the 1994 ACM conference on Computer supported cooperative work, 1994, pp. 175–186.
[18] U. Shardanand and P. Maes, “Social information filtering: algorithms for automating ‘word of mouth,’” in Proceedings of the ACM Conference on Human Factors in Computing Systems, 1995, vol. 1, pp. 210–217.
[19] W. Hill, W. Hill, L. Stead, L. Stead, M. Rosenstein, M. Rosenstein, G. Furnas, and G. Furnas, “Recommending and evaluating choices in a virtual community of use,” in Proceedings of the SIGCHI conference on Human factors in computing systems, 1995, pp. 194–201.
[20] J. S. Breese, D. Heckerman, and C. Kadie, “Empirical analysis of predictive algorithms for collaborative filtering,” in UAI’98 Proceedings of the Fourteenth conference on Uncertainty in artificial intelligence, 1998, pp. 43–52.