Filters








93,527 Hits in 6.8 sec

Evaluating information retrieval system performance based on user preference

Bing Zhou, Yiyu Yao
2009 Journal of Intelligent Information Systems  
We find that the normalized distance performance measure is a good choice in terms of the sensitivity to document rank order and gives higher credits to systems for their ability to retrieve highly relevant  ...  The main objective of this paper is to propose a framework for system evaluation based on user preference of documents.  ...  Ras during the ISMIS 2008 conference in Toronto, and for the valuable suggestions from anonymous reviewers.  ... 
doi:10.1007/s10844-009-0096-5 fatcat:i5qrpg3acnaq3ln3fspl7svsvq

Active learning based clothing image recommendation with implicit user preferences

Chiao-Meng Huang, Chia-Po Wei, Yu-Chiang Frank Wang
2013 2013 IEEE International Conference on Multimedia and Expo Workshops (ICMEW)  
With a recently developed sparse-coding based algorithm for content-based image retrieval, we utilize support vector regression (SVR) with a user-interaction training stage to observe user preferences  ...  Different from prior retrieval approaches, we advance an active learning scheme during retrieval for inferring user preferences.  ...  Acknowledgement This work is supported in part by National Science Council of Taiwan via NSC100-2221-E-001-018-MY2.  ... 
doi:10.1109/icmew.2013.6618318 dblp:conf/icmcs/HuangWW13 fatcat:jgemkr6ux5dkpegifgo5v6ce7u

Personalized Music Recommendation with Triplet Network [article]

Haoting Liang, Donghuo Zeng, Yi Yu, Keizo Oyama
2019 arXiv   pre-print
Some common problems in recommendation system like feature representations, distance measure and cold start problems are also challenges for music recommendation.  ...  In this paper, I proposed a triplet neural network, exploiting both positive and negative samples to learn the representation and distance measure between users and items, to solve the recommendation task  ...  D() is a distance function to measure the distance between users and items in the latent common space.  ... 
arXiv:1908.03738v1 fatcat:4pyovzdqvvgivcugp5csstyeki

Experiments in CLIR using fuzzy string search based on surface similarity

Sethuramalingam Subramaniam, Anil Kumar Singh, Pradeep Dasigi, Vasudeva Varma
2009 Proceedings of the 32nd international ACM SIGIR conference on Research and development in information retrieval - SIGIR '09  
We found significant improvements for all the six language pairs using a method for fuzzy text search based on Surface Similarity.  ...  In this paper we report these results and compare them with a baseline CLIR system and a CLIR system that uses Scaled Edit Distance (SED) for fuzzy string matching.  ...  The reason for preferring SED over the normal edit distance is that it reduces the disparities between long and short words. The comparison of results is shown in Table 2 .  ... 
doi:10.1145/1571941.1572076 dblp:conf/sigir/SubramaniamSDV09 fatcat:lrn7qvqb4fgynkgubuepbc7q5e

Ranking-oriented nearest-neighbor based method for automatic image annotation

Chaoran Cui, Jun Ma, Tao Lian, Xiaofang Wang, Zhaochun Ren
2013 Proceedings of the 36th international ACM SIGIR conference on Research and development in information retrieval - SIGIR '13  
Automatic image annotation plays a critical role in keywordbased image retrieval systems.  ...  In particular, a new learning to rank algorithm is developed, which exploits the implicit preference information of labeled images and underlines the accuracy of the top-ranked results.  ...  To allow encoding this kind of information, in this paper, we construct the preference information in the form of ordered pairs of images, and also assign each pair a weight to represent the importance  ... 
doi:10.1145/2484028.2484113 dblp:conf/sigir/CuiMLWR13 fatcat:f7us5ca22vhexjeogrbyvpqayy

Heterogeneous Metric Learning for Cross-Modal Multimedia Retrieval [chapter]

Jun Deng, Liang Du, Yi-Dong Shen
2013 Lecture Notes in Computer Science  
In this paper, we propose a Bayesian personalized ranking based heterogeneous metric learning (BPRHML) algorithm, which optimizes for correctly ranking the retrieval results.  ...  Due to the massive explosion of multimedia content on the web, users demand a new type of information retrieval, called cross-modal multimedia retrieval where users submit queries of one media type and  ...  We would like to thank all anonymous reviewers for their helpful comments. This work is supported in part by NSFC grant 60970045 and China National 973 project 2013CB329305.  ... 
doi:10.1007/978-3-642-41230-1_4 fatcat:xtfbwdmwufbqzkf2s36t53jxm4

Deep Feature Learning with Manifold Embedding for Robust Image Retrieval

Xin Chen, Ying Li
2020 Algorithms  
Thus, we maintain both the efficiency of Euclidean distance-based similarity measurement and the effectiveness of manifold information in the new feature embedding.  ...  Conventionally, the similarity between two images is measured by the easy-calculating Euclidean distance between their corresponding image feature representations for image retrieval.  ...  Based on this, we can extract new image features for database images, calculate a new distance matrix and a refined affinity matrix, mine new manifold information and training pairs, and learn a new feature  ... 
doi:10.3390/a13120318 fatcat:s4tqqlk3kzcahozf6pyhhg6c6u

Active Exploration of Large 3D Model Repositories

Lin Gao, Yan-Pei Cao, Yu-Kun Lai, Hao-Zhi Huang, Leif Kobbelt, Shi-Min Hu
2015 IEEE Transactions on Visualization and Computer Graphics  
Existing model retrieval techniques do not scale well with the size of the database since often a large number of very similar objects are returned for a query, and the possibilities to refine the search  ...  We achieve this by an offline pre-processing stage, where global and local shape descriptors are computed for each model and a sparse distance metric is derived that can be evaluated efficiently even for  ...  Global active learning is used to improve retrieved models based on whole models whereas local active learning is used for using exploiting local preference information.  ... 
doi:10.1109/tvcg.2014.2369039 pmid:26529460 fatcat:mekpvw5zmffbfmsmjzzudgx4ti

Development of Fashion Product Retrieval and Recommendations Model Based on Deep Learning

Jaechoon Jo, Seolhwa Lee, Chanhee Lee, Dongyub Lee, Heuiseok Lim
2020 Electronics  
However, the text-based search method has limitations because of the nature of the fashion industry, in which design is a very important factor.  ...  Therefore, we developed an intelligent fashion technique based on deep learning for efficient fashion product searches and recommendations consisting of a Sketch-Product fashion retrieval model and vector-based  ...  Acknowledgments: The authors thank the MSIT, IITP, and NRF for supporting the research and project. Conflicts of Interest: The authors declare no conflicts of interest.  ... 
doi:10.3390/electronics9030508 fatcat:kmffxze2zbemfhqwxcioksveui

A Novel Alignment-Free Method for Comparing Transcription Factor Binding Site Motifs

Minli Xu, Zhengchang Su, Vladimir Brusic
2010 PLoS ONE  
The majority of current methods for motif comparison involve a similarity metric for column-to-column comparison and a method to find the optimal position alignment between the two compared motifs.  ...  We often need to compare TFBS motifs using their PFMs in order to search for similar motifs in a motif database, or cluster motifs according to their binding preference.  ...  Acknowledgments We would like to thank Miss Shan Li for her critical reading of this manuscript and constructive suggestions.  ... 
doi:10.1371/journal.pone.0008797 pmid:20098703 pmcid:PMC2808352 fatcat:ooppl674wnap5eoghs4ezyetzi

Semi-supervised learning to rank with preference regularization

Martin Szummer, Emine Yilmaz
2011 Proceedings of the 20th ACM international conference on Information and knowledge management - CIKM '11  
The regularizer captures manifold structure in the data, and we also propose a rank-sensitive version designed for top-heavy retrieval metrics including NDCG and mean average precision.  ...  We introduce a preference regularizer favoring that similar items are similar in preference to each other.  ...  Acknowledgements We thank Massih Reza Amini for making the semi-supervised RankBoost [2] implementation available, and Tom Minka and Vishwa Vinay for helpful discussions.  ... 
doi:10.1145/2063576.2063620 dblp:conf/cikm/SzummerY11 fatcat:yrr5hyxn2zel3jhjp3gbtnbyxa

Learning with Batch-wise Optimal Transport Loss for 3D Shape Recognition [article]

Lin Xu, Han Sun, Yuai Liu
2019 arXiv   pre-print
The widely used pair-wise (or triplet) based loss objectives cannot make full use of semantical information in training samples or give enough attention to those hard samples during optimization.  ...  We use it to learn the distance metric and deep feature representation jointly for recognition.  ...  Sketch-based 3D Shape Retrieval We then evaluated our method for sketch-based 3D shape retrieval on two large-scale benchmark datasets, i.e., SHREC13 and SHREC14.  ... 
arXiv:1903.08923v1 fatcat:szahnuma45g2znkg4teqw5aavm

Content-Based Video-Music Retrieval Using Soft Intra-Modal Structure Constraint [article]

Sungeun Hong, Woobin Im, Hyun S. Yang
2017 arXiv   pre-print
This paper introduces a new content-based, cross-modal retrieval method for video and music that is implemented through deep neural networks.  ...  Up to now, only limited research has been conducted on cross-modal retrieval of suitable music for a specified video or vice versa.  ...  Comparison with other methods. Table III presents comparisons of our VM-NET with previous work.  ... 
arXiv:1704.06761v2 fatcat:rmno6ua55bhhnbz3kxzlt7ktge

Scene Parsing with Global Context Embedding [article]

Wei-Chih Hung, Yi-Hsuan Tsai, Xiaohui Shen, Zhe Lin, Kalyan Sunkavalli, Xin Lu, Ming-Hsuan Yang
2017 arXiv   pre-print
We present a scene parsing method that utilizes global context information based on both the parametric and non- parametric models.  ...  Compared to previous methods that only exploit the local relationship between objects, we train a context network based on scene similarities to generate feature representations for global contexts.  ...  Acknowledgments This work is supported in part by the NSF CAREER Grant #1149783, gifts from Adobe and NVIDIA.  ... 
arXiv:1710.06507v2 fatcat:rpnpodk6mzekxgn3mw2uxm5a6e

Learning With Batch-Wise Optimal Transport Loss for 3D Shape Recognition

Lin Xu, Han Sun, Yuai Liu
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
The widely used pair-wise (or triplet) based loss objectives cannot make full use of semantical information in training samples or give enough attention to those hard samples during optimization.  ...  We use it to learn the distance metric and deep feature representation jointly for recognition.  ...  (b): Only the semantical information of a pair of examples is considered at each update in the pair-wise case.  ... 
doi:10.1109/cvpr.2019.00345 dblp:conf/cvpr/XuSL19 fatcat:yms72ja3djfurctwhpexdg35aq
« Previous Showing results 1 — 15 out of 93,527 results