The use of MMR, diversity-based reranking for reordering documents and producing summaries
Proceedings of the 21st annual international ACM SIGIR conference on Research and development in information retrieval
Computers and Intractability: A Guide to the Theory of NP-Completeness
Computers and Intractability: A Guide to the Theory of NP-Completeness
Beyond independent relevance: methods and evaluation metrics for subtopic retrieval
Proceedings of the 26th annual international ACM SIGIR conference on Research and development in informaion retrieval
Introduction to MPEG-7: Multimedia Content Description Interface
Introduction to MPEG-7: Multimedia Content Description Interface
Distinctive Image Features from Scale-Invariant Keypoints
International Journal of Computer Vision
Updating pagerank with iterative aggregation
Proceedings of the 13th international World Wide Web conference on Alternate track papers & posters
Multimodal concept-dependent active learning for image retrieval
Proceedings of the 12th annual ACM international conference on Multimedia
CVPRW '04 Proceedings of the 2004 Conference on Computer Vision and Pattern Recognition Workshop (CVPRW'04) Volume 12 - Volume 12
Improving recommendation lists through topic diversification
WWW '05 Proceedings of the 14th international conference on World Wide Web
A Comparison of Affine Region Detectors
International Journal of Computer Vision
Less is more: probabilistic models for retrieving fewer relevant documents
SIGIR '06 Proceedings of the 29th annual international ACM SIGIR conference on Research and development in information retrieval
Diversifying the image retrieval results
MULTIMEDIA '06 Proceedings of the 14th annual ACM international conference on Multimedia
Pagerank for product image search
Proceedings of the 17th international conference on World Wide Web
Novelty and diversity in information retrieval evaluation
Proceedings of the 31st annual international ACM SIGIR conference on Research and development in information retrieval
Introduction to Information Retrieval
Introduction to Information Retrieval
CARD: a decision-guidance framework and application for recommending composite alternatives
Proceedings of the 2008 ACM conference on Recommender systems
Proceedings of the Second ACM International Conference on Web Search and Data Mining
An axiomatic approach for result diversification
Proceedings of the 18th international conference on World wide web
CrowdReranking: exploring multiple search engines for visual search reranking
Proceedings of the 32nd international ACM SIGIR conference on Research and development in information retrieval
Jointly optimising relevance and diversity in image retrieval
Proceedings of the ACM International Conference on Image and Video Retrieval
Structured search result differentiation
Proceedings of the VLDB Endowment
Overview of the ImageCLEFphoto 2008 photographic retrieval task
CLEF'08 Proceedings of the 9th Cross-language evaluation forum conference on Evaluating systems for multilingual and multimodal information access
DivRank: the interplay of prestige and diversity in information networks
Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining
Evaluating Color Descriptors for Object and Scene Recognition
IEEE Transactions on Pattern Analysis and Machine Intelligence
ACM SIGMOD Record
Fast incremental and personalized PageRank
Proceedings of the VLDB Endowment
Incremental diversification for very large sets: a streaming-based approach
Proceedings of the 34th international ACM SIGIR conference on Research and development in Information Retrieval
Diversified ranking on large graphs: an optimization viewpoint
Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining
Max-Sum diversification, monotone submodular functions and dynamic updates
PODS '12 Proceedings of the 31st symposium on Principles of Database Systems
Maximizing a Monotone Submodular Function Subject to a Matroid Constraint
SIAM Journal on Computing
Hi-index | 0.01 |
Different from conventional retrieval methods that focus on presenting users with relevant results, diversified ranking algorithms aim to provide a non-redundant and rich topic coverage of information in the top ranked list. Most existing approaches for diversified image ranking require a two-step process: (1) a set of potentially relevant images is determined; (2) these images are re-ranked to be diverse. However, the diversified ranking lists obtained from these methods are under the uniform constraint, which treats each image equally important. As such, the final diversified ranking list may come from one or several limited number of categories, where they are the most relevant to the query, rather than exhibiting a wide range of categories to the users. Motivated by this, in this paper, we propose a novel diversified ranking algorithm on image retrieval associated with a non-uniform matroid constraint (each image is weighted differently according to the category it belongs to, and the partition matroid constraints are posed on each category) to leverage the relevance and diversity in an optimized fashion, which jointly captures the relevance and diversity. We introduce hardness coefficient that can automatically set the weight parameters regarding each image to meet the requirement of non-uniform matroid partition constraints. We study our method regarding both in-sample queries and out-of-sample queries. For in-sample queries, we prove that maximizing the proposed objective function is NP-hard. As a result, we find a near-optimal solution with a guaranteed approximation ratio. To tackle the case of out-of-sample queries, four strategies are investigated. Extensive experimental evaluations on real image data sets demonstrate the effectiveness and efficiency of our method.