184,971 Hits in 6.9 sec

A Group Testing Framework for Similarity Search in High-dimensional Spaces

Miaojing Shi, Teddy Furon, Hervé Jégou
2014 Proceedings of the ACM International Conference on Multimedia - MM '14  
This paper introduces a group testing framework for detecting large similarities between high-dimensional vectors, such as descriptors used in state-of-the-art description of multimedia documents.  ...  Unlike concurrent indexing methods that suffer from the curse of dimensionality, our method exploits the properties of high-dimensional spaces.  ...  CONCLUSION Our framework for similarity search in high-dimensional spaces amounts to replacing the individual query-database similarity computations by group measurements, in a way strongly inspired by  ... 
doi:10.1145/2647868.2654895 dblp:conf/mm/ShiFJ14 fatcat:zavp7imecrewtpsxlewixov3oi

Applying Z-Curve Technique to Compute Skyline Set in Multi Criteria Decision Making System

T. Vijaya Saradhi, Kodukula Subrahmanyam, P. Venkateswara Rao, Hye-jin Kim
2016 International Journal of Database Theory and Application  
In skyline computation major cost depends on finding dominance tests between high dimensional objects and the order in which they are accessing.  ...  Space filling Z-curve is the best suitable way to address the challenges in skyline computation.  ...  Conclusion In this work, we proposed a novel method for organizing and retrieving of high dimensional data to enable skyline computation using Z-order curve.  ... 
doi:10.14257/ijdta.2016.9.12.02 fatcat:6bkk7yy6ezedjfan7nc46zogru

Scaling Group Testing Similarity Search

Ahmet Iscen, Laurent Amsaleg, Teddy Furon
2016 Proceedings of the 2016 ACM on International Conference on Multimedia Retrieval - ICMR '16  
However, similarity search techniques inspired by the group testing framework have recently been proposed in an attempt to specifically defeat the curse of dimensionality.  ...  This paper identifies these difficulties and proposes extensions to the group testing framework for similarity searches that allow to handle larger collections of feature vectors.  ...  Similarity search techniques inspired by the group testing framework have recently been proposed in an attempt to specifically defeat the curse of dimensionality.  ... 
doi:10.1145/2911996.2912010 dblp:conf/mir/IscenAF16 fatcat:65jj6tyrdjgjlhxyub6bud2iia

BrePartition: Optimized High-Dimensional kNN Search with Bregman Distances [article]

Yang Song, Yu Gu, Rui Zhang, Ge Yu
2020 arXiv   pre-print
This paper addresses the urgent problem of high-dimensional kNN search with Bregman distances. We propose a novel partition-filter-refinement framework.  ...  Such high-dimensional space has posed significant challenges for existing kNN search algorithms with Bregman distances, which could only handle data of medium dimensionality (typically less than 100).  ...  OVERALL FRAMEWORK In this paper, we solve the kNN search problem with Bregman distances in high-dimensional space by our proposed partition-filter-refinement framework.  ... 
arXiv:2006.00227v1 fatcat:cvuizn6xbjebze2q77sp2x2vce

Efficient Semantic-Based Content Search in P2P Network

Heng Tao Shen, Yanfeng Shu, Bei Yu
2004 IEEE Transactions on Knowledge and Data Engineering  
First, we propose a general and extensible framework for searching similar documents in P2P network. The framework is based on the novel concept of Hierarchical Summary Structure.  ...  In this paper, we present the design of a distributed P2P information sharing system that supports semantic-based content searches of relevant documents.  ...  As the network size grows, efficient searching in high-dimensional space becomes prevalently important.  ... 
doi:10.1109/tkde.2004.1318564 fatcat:y6ldhgvplvdnfpcj6oqlu4ycqa


Shilpa R .
2016 International Journal of Research in Engineering and Technology  
An image retrieval system for a large database of digital images is a computer system for searching, retrieving and browsing image and obtaining most similar images.  ...  For most hashing strategies, the performance of retrieval vigorously relies upon the decision of the high-dimensional component descriptor.  ...  INTRODUCTION Hashing is qualified and most prevalent technique in huge scale database for nearest (closest )neighbor search inserting high-dimensional component descriptors into a similarity preserving  ... 
doi:10.15623/ijret.2016.0516020 fatcat:wndeunahcrghjnq5inlw2tvigq

Feature and Search Space Reduction for Label-Dependent Multi-label Classification [chapter]

Prema Nedungadi, H. Haripriya
2015 Advances in Intelligent Systems and Computing  
A strategy is proposed to combine both multiple regression and hybrid k-Nearest Neighbor algorithm in an efficient way for high-dimensional multi-label classification.  ...  The hybrid kNN performs the dimensionality reduction in the feature space of multi-labeled data in order to reduce the search space as well as the feature space for kNN, and multiple regression is used  ...  Conclusion Multi-label data classification in high-dimensional space is a new area to explore. In our paper, we proposed a new method for high-dimensional multi-label prediction.  ... 
doi:10.1007/978-81-322-2523-2_57 fatcat:ksvzr2h7gbew5kxs27wfc54mki

Sparse hashing for fast multimedia search

Xiaofeng Zhu, Zi Huang, Hong Cheng, Jiangtao Cui, Heng Tao Shen
2013 ACM Transactions on Information Systems  
Hash-based methods achieve fast similarity search by representing high-dimensional data with compact binary codes.  ...  In this article, we focus on these tasks to implement approximate similarity search by proposing a novel hash based method named sparse hashing (SH for short).  ...  CONCLUSION The article proposed a novel sparse hashing framework for fast approximate similarity search.  ... 
doi:10.1145/2457465.2457469 fatcat:sawxwotnyzg2xeh7pu74gy4z6q

CrossMedia: Supporting Collaborative Research of Media Retrieval

Péter Mátételki, László Havasi, Márton Gergó, András Micsik, Ákos Kiss, Tamás Szirányi, László Kovács
2013 Procedia - Social and Behavioral Sciences  
In the paper a flexible framework for research purposes is introduced for testing features, metrics, distances and indexing structures.  ...  Additionally, we compare LHI-tree to FLANN, an effective implementation of ANN search and show that LHI-tree gives similar list of retrieved images.  ...  A recent method proposed in [5] is the Nearest Vector Tree which is designed for approximate nearest neighbor search in very large, high-dimensional databases.  ... 
doi:10.1016/j.sbspro.2013.02.083 fatcat:ogahexeth5ai7gstqgdaz4lz7u

Leveraging Reinforcement Learning for evaluating Robustness of KNN Search Algorithms [article]

Pramod Vadiraja, Christoph Peter Balada
2021 arXiv   pre-print
In very high dimensional spaces the K-nearest neighbor search (KNNS) suffers in terms of complexity in computation of high dimensional distances.  ...  Finally, in order to evaluate the robustness of a KNNS approach against adversarial points, we propose a generic Reinforcement Learning based framework for the same.  ...  high dimensional spaces.  ... 
arXiv:2102.06525v1 fatcat:d5tfa2nuknbzzcld42satedhiy

Discovering salient characteristics of authors of artworks

Peter Bajcsy, Maryam Moslemi, David G. Stork, Jim Coddington, Anna Bentkowska-Kafel
2010 Computer Vision and Image Analysis of Art  
We accomplished this by exploring a large search space of low level image descriptors.  ...  By employing our framework we had not only saved time of art historians but also provided quantitative measures for incorporating their personal judgments and bridging the semantic gap in image understanding  ...  Specific Framework Following the general methodology, we focused on investigating a specific framework in which one could discover discriminating characteristics of artists in a large dimensional space  ... 
doi:10.1117/12.838847 fatcat:hkowvbz32jfgnah2az7gmplbca

Non-redundant Multi-view Clustering via Orthogonalization

Ying Cui, Xiaoli Z. Fern, Jennifer G. Dy
2007 Seventh IEEE International Conference on Data Mining (ICDM 2007)  
We test our framework on both synthetic and high-dimensional benchmark data sets, and the results show that indeed our approaches were able to discover varied solutions that are interesting and meaningful  ...  This is especially true for high-dimensional data, where different feature subspaces may reveal different structures of the data.  ...  An integral part of our framework is to search a clustering structure in a high-dimensional space and find the corresponding subspace that best reveals the clustering structure.  ... 
doi:10.1109/icdm.2007.94 dblp:conf/icdm/CuiFD07 fatcat:7aw4ava33rgy3jjj62jnfzt5o4

Entropy-Scaling Search of Massive Biological Data

Y. William Yu, Noah M. Daniels, David Christian Danko, Bonnie Berger
2015 Cell Systems  
Here, we introduce a framework for similarity search based on characterizing a dataset's entropy and fractal dimension.  ...  Using these ideas, we present accelerated versions of standard tools, with no loss in specificity and little loss in sensitivity, for use in three domains---high-throughput drug screening (Ammolite, 150x  ...  Barrile for graphic design, Simon Ye for providing a bug fix to MICA, and Jian Peng for suggesting high-throughput drug screening as an application.  ... 
doi:10.1016/j.cels.2015.08.004 pmid:26436140 pmcid:PMC4591002 fatcat:57hcnc6p3zax5kfgudwj5wsscy

Mining Mass Spectra: Metric Embeddings and Fast Near Neighbor Search [article]

Debojyoti Dutta, Ting Chen
2006 arXiv   pre-print
We first robustly embed spectra in a high dimensional space in a novel fashion and then apply fast approximate near neighbor algorithms for tasks such as constructing filters for database search, indexing  ...  In this paper, we present a general framework based on vector spaces to avoid pair-wise comparisons.  ...  Once we have embedded the spectra in a Euclidean space, we can use some of the common techniques to visualize high dimensional data by dimensionality reduction.  ... 
arXiv:q-bio/0603002v1 fatcat:s3jzq24tjnbuvnszg23mjdc4fi

Unconventional application of k-means for distributed approximate similarity search [article]

Felipe Ortega and Maria Jesus Algar and Isaac Martín de Diego and Javier M. Moguerza
2022 arXiv   pre-print
Similarity search based on a distance function in metric spaces is a fundamental problem for many applications.  ...  An implementation of this new indexing method is evaluated, using a synthetic dataset and a real-world dataset in a high-dimensional and high-sparsity space.  ...  Metric Access Methods (MAM) provide a more general framework for similarity search problems set out in metric spaces [1, 3] .  ... 
arXiv:2208.02734v1 fatcat:hylpglx3knbvdgqlqdrsxx7xce
« Previous Showing results 1 — 15 out of 184,971 results