Schein, AI, Popescul, A and Ungar, LH, 2002, Methods and metrics for cold-start recommendations. In Proceedings of the 25th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 253--260. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Computing Conditional Probabilities In Large Domains By Maximizing.. - III (2003) (1 citation) (Correct)
....User recommendations using (a) the most similar images and (b) the most similar images assuming the user doesn t like the previous recommendations. 5.3. 2 Overcoming the Cold Start Problem using CBIR A common problem for collaborative filtering algorithms such as ours is the cold start problem [52, 70]. That is, our algorithm needs a large amount of training data from users before it can start producing good results. Unfortunately, users won t use the system until it produces good results. We see two methods for solving this problem. First, content or keyword based image retrieval systems could ....
Andrew I. Schein, Alexandrin Popescul, and Lyle H. Ungar. Methods and metrics for cold-start recommendations. In Proceedings of the 25'th annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 2002.
Learning User Profiles for Content-Based Filtering in .. - Abbattista.. (2002) (2 citations) (Correct)
....for whom little or no information has yet been acquired. In fact, to be able to make accurate predictions, the system must first learn the user preferences from the ratings that she makes. If the system does not show quick progress, a user may lose patience and stop using the system. Schein et al. [21] propose a probabilistic model that combines content and collaborative information by using expectation maximization learning to fit the model to the data. Another recent approach [11] exploits ontologies to investigate how domain knowledge can help in the acquisition of user preferences. ....
Schein A. I., Popescul A., Ungar L. H., Methods and Metrics for Cold-Start Recommendations, Proceedings of the XXV Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, Tampere, Finland (2002).
CROC: A New Evaluation Criterion for Recommender Systems - Andrew Schein Alexandrin Self-citation (Schein Popescul Ungar Pennock) (Correct)
No context found.
A. I. Schein, A. Popescul, L. H. Ungar, and D. M. Pennock. Methods and metrics for cold-start recommendations. In Proceedings of the 25'th International ACM Conference on Research and Development in Information Retrieval, 2002.
Structural Logistic Regression for Link Analysis - Popescul, Ungar (2003) Self-citation (Popescul Ungar) (Correct)
No context found.
Andrew I. Schein, Alexandrin Popescul, Lyle H. Ungar, and David M. Pennock. Methods and metrics for cold-start recommendations. In Proc. of the 25'th Conference on Research and Development in Information Retreival (SIGIR 2002), 2002.
Structural Logistic Regression for Link Analysis - Popescul, Ungar (2003) Self-citation (Popescul Ungar) (Correct)
No context found.
Andrew I. Schein, Alexandrin Popescul, Lyle H. Ungar, and David M. Pennock. Methods and metrics for cold-start recommendations. In Proc. of the 25'th Conference on Research and Development in Information Retreival (SIGIR 2002), 2002.
Statistical Relational Learning for Link Prediction - Popescul, Ungar (2003) (2 citations) Self-citation (Popescul Ungar) (Correct)
....of citations can be recalled for recommendation with 91 precision. This is an overall measure of performance some users can receive more than enough recommendations, and others none. When we want to recommend a fixed number of citations to every user, the CROC performance metric should be used [Schein et al. 2002] . 5 Related Work Integrating or extending existing models or techniques to relational data has been addressed by researchers in several fields, including inductive logic programming, belief nets and link analysis. A number of approaches extending one table learners to multi table domains have ....
Andrew I. Schein, Alexandrin Popescul, Lyle H. Ungar, and David M. Pennock. Methods and metrics for cold-start recommendations. In Proceedings of the 25'th International ACM Conference on Research and Development in Information Retreival (SIGIR 2002), 2002.
Notes on the CROC Curve - Schein Self-citation (Schein) (Correct)
....by recommending to the most active users. This can lead to a misleading sense of performance for applications where coverage of the user base has importance. We developed the CROC for use in conjunction with the GROC curve to explore this issue of user coverage in evalulation. In previous work [3] we show that the two measures can have little predictive power over each other making the combined analysis much more insightful than the two pieces independently. Like the GROC curve, the CROC curve is built from the same definition of hit rate and false alarm rate. However, we constrain the ....
....has a di#erent (but usually overlapping) set of items to rate in the test set. However, we can not recommend k movies to p if p only has k # k items to rate in the test set. So we recommend a maximum of k # items to this user. In cold start ratings imputation and ratings prediction evaluation [3] each user has the same number of item observations in the testing phase. We also find that each user has the same number of test set observations for applications with repeated observations allowed (i.e. we do not eliminate person item pairs that occur in training data) For cases where each user ....
Andrew Ian Schein, Alexandrin Popescul, Lyle H. Ungar, and David M. Pennock. Methods and metrics for cold-start recommendations. In Proceedings of the 25'th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 2002.
REFEREE: An open framework for practical testing of.. - Cosley, Lawrence.. (2002) (4 citations) Self-citation (Pennock) (Correct)
....with domain specific descriptive information about the items after all, there is no reason to throw the bagof words out with the trash. Using content information is a promising way to overcome sparsity and the newitem and new user problems (sometimes referred to as the cold start problem [23]) A serious impediment to progress in recommender systems research is the lack of a standard framework for evaluating competing algorithms in a real world setting. We argue that ResearchIndex can serve as an excellent testbed for recommendation algorithms, especially hybrid algorithms that ....
SCHEIN, A. I., POPESCUL, A., UNGAR, L. H., AND PENNOCK, D. M. Methods and metrics for cold-start recommendations. To appear in Proceedings of the 25th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2002).
User Profiling in Personal . . . - Godoy, al. (2005) (Correct)
No context found.
Schein, AI, Popescul, A and Ungar, LH, 2002, Methods and metrics for cold-start recommendations. In Proceedings of the 25th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 253--260.
A Graph Model for E-Commerce Recommender Systems - Huang, Chung, Chen (2004) (Correct)
No context found.
Schein, A.I., Popescul, A., Unger, L.H., & Pennock, D.M. (2002). Methods and metrics for cold-start recommendations. In Proceedings of the 25th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2002) (pp. 253--260), Tampere, Finland.
Personalized Information Access in a Bibliographic.. - Haase, Ehrig, Hotho, .. (2004) (Correct)
No context found.
Schein, A.; Popescul, A.; Ungar, L.; and Pennock, D. 2002. Methods and metrics for cold-start recommendations. In Proceedings of the 25th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval.
The Organisation and Retrieval of Document Collections: A.. - Vinokourov (2003) (Correct)
No context found.
A. Schein, A. Popescul, L. Ungar & D. Pennock. Methods and metrics for cold-start recommendations. In Proceedings of the 25th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 2002. 185
Coupling Niche Browsers and Affect Analysis for an.. - Application Gregory.. (Correct)
No context found.
Schein A. I., Popescul, A., and Ungar, L. H. (2002). Methods and Metrics for Cold-Start Recommendations. In Proceedings of the XXV Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, Tampere, Finland.
Trust-aware Decentralized Recommender Systems: PhD Research Proposal - Massa (2003) (Correct)
No context found.
A. Schein, A. Popescul, L. Ungar, and D. Pennock. Methods and metrics for cold-start recommendations, 2002.
Trust-aware Decentralized Recommender Systems: PhD Research Proposal - Massa (2004) (Correct)
No context found.
A. Schein, A. Popescul, L. Ungar, and D. Pennock. Methods and metrics for cold-start recommendations, 2002.
Generating Personalized Recommendations in a.. - Stegmann, Koch.. (2003) (Correct)
No context found.
A. Schein, A. Popescul, L. Ungar, and D. Pennock. Methods and Metrics for Cold-Start Recommendations. In Proceedings of the 25th ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2002).
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC