Filters








140,663 Hits in 6.0 sec

Distance-based indexing for high-dimensional metric spaces

Tolga Bozkaya, Meral Ozsoyoglu
1997 Proceedings of the 1997 ACM SIGMOD international conference on Management of data - SIGMOD '97  
In this paper, we introduce a distance based index structure called multi-vantage point (mvp) tree for similarity queries on high-dimensional metric spaces.  ...  Distance based index structures are proposed for applications where the data domain is high dimensional, or the distance function used to compute distances between data objects is non-Euclidean.  ...  Here, we are mainly concerned on distance based indexing for high-dimensional metric spaces. We also concentrate on the near neighbor queries when we introduce our index structure.  ... 
doi:10.1145/253260.253345 dblp:conf/sigmod/BozkayaO97 fatcat:7zqybhoebfb5rnlo4cjglcdeze

Distance-based indexing for high-dimensional metric spaces

Tolga Bozkaya, Meral Ozsoyoglu
1997 SIGMOD record  
In this paper, we introduce a distance based index structure called multi-vantage point (mvp) tree for similarity queries on high-dimensional metric spaces.  ...  Distance based index structures are proposed for applications where the data domain is high dimensional, or the distance function used to compute distances between data objects is non-Euclidean.  ...  Here, we are mainly concerned on distance based indexing for high-dimensional metric spaces. We also concentrate on the near neighbor queries when we introduce our index structure.  ... 
doi:10.1145/253262.253345 fatcat:xwwra6arz5fe5ayx2tnavxw2n4

Indexing Issues in Supporting Similarity Searching [chapter]

Hanan Samet
2004 Lecture Notes in Computer Science  
This includes a discussion of the curse of dimensionality, as well as multidimensional indexing, distance-based indexing, dimension reduction, and embedding methods.  ...  Indexing issues that arise in the support of similarity searching are presented.  ...  Concluding Remarks Providing indexing support for similarity searching is an important area where much work remains to be done.  ... 
doi:10.1007/978-3-540-30542-2_57 fatcat:7nysehgscvet5asn7oho22towy

Techniques for similarity searching in multimedia databases

Hanan Samet
2010 Proceedings of the VLDB Endowment  
This includes a discussion of the curse of dimensionality, as well as multidimensional indexing, distancebased indexing, and the actual search process which is realized by nearest neighbor finding.  ...  Techniques for similarity searching in multimedia databases are reviewed.  ...  Searching in high-dimensional spaces is time-consuming.  ... 
doi:10.14778/1920841.1921064 fatcat:z63jjl5gabd5ddbocbppbl5h3i

Parameterized earth mover's distance for efficient metric space indexing

Jakub Lokoč, Christian Beecks, Thomas Seidl, Tomáš Skopal
2011 Proceedings of the Fourth International Conference on SImilarity Search and APplications - SISAP '11  
We empirically show, that we can significantly improve the indexability of the distance space and that we can tune the retrieval quality by adapting the parameterized Earth Mover's Distance.  ...  the indexability.  ...  However, the distance spaces based on the Earth Mover's Distance usually suffer from high intrinsic dimensionality and thus only the approximate search can speedup query processing by metric index.  ... 
doi:10.1145/1995412.1995438 dblp:conf/sisap/LokocBSS11 fatcat:g6atnpvfbzar3prquiv2n4iyly

Case Study: Distance-Based Image Retrieval in the MoBIoS DBMS

Rui Mao, Wenguo Liu, D.P. Miranker, Q. Iqbal
2005 The Fifth International Conference on Computer and Information Technology (CIT'05)  
It has been shown that for high dimensional uniform vectors with similarity norms, any clustering and partitioning index method is outperformed by sequential scan.  ...  Similarity search leveraging distance-based index structures is increasingly being used for complex data types.  ...  Its intrinsic clustering usually leads to low intrinsic dimensionality although the feature vectors are in high dimension space, suggesting the application of distance-based indexing, or metric-space indexing  ... 
doi:10.1109/cit.2005.83 dblp:conf/IEEEcit/MaoLMI05 fatcat:4uevp65agnde3mrkbvxqlkhjja

A Parallel Similarity Search in High Dimensional Metric Space Using M-Tree [chapter]

Adil Alpkocak, Taner Danisman, Ulker Tuba
2002 Lecture Notes in Computer Science  
I n this study, parallel implementation of M -tree to index high dimensional metric space has been elaborated and an optimal declustering technique has been proposed.  ...  First, we have defined the optimal declustering and developed an algorithm based on this definition.  ...  high dimensional metric space.  ... 
doi:10.1007/3-540-47840-x_16 fatcat:tkp4a6otn5cotmznxllk7xx5ii

Indexing the signature quadratic form distance for efficient content-based multimedia retrieval

Christian Beecks, Jakub Lokoč, Thomas Seidl, Tomáš Skopal
2011 Proceedings of the 1st ACM International Conference on Multimedia Retrieval - ICMR '11  
In this paper, we investigate the indexability of the Signature Quadratic Form Distance regarding metric access methods.  ...  Although the Signature Quadratic Form Distance has shown good retrieval performance with respect to their qualities of effectiveness and efficiency, its applicability to index structures remains a challenging  ...  The responsibility for the content of this publication lies with the authors.  ... 
doi:10.1145/1991996.1992020 dblp:conf/mir/BeecksLSS11 fatcat:qusznahycbck7k5s5ynal32fpq

Dimension reduction for distance-based indexing

Rui Mao, Willard L. Miranker, Daniel P. Miranker
2010 Proceedings of the Third International Conference on SImilarity Search and APplications - SISAP '10  
dimensions • Metric space indexing vs. high dimensional indexing 3.  ...  reduction for distance-based indexing • PCA for distance-based indexing • Empirical results • Conclusions and future work General steps of Distance-based Indexing 1. metric space  R k 2. multi-dimensional  ... 
doi:10.1145/1862344.1862349 dblp:conf/sisap/MaoMM10 fatcat:b5cuotf5zzfwhbind7hxzdu3hm

An Indexing Approach for Representing Multimedia Objects in High-Dimensional Spaces Based on Expectation Maximization Algorithm [chapter]

Giuseppe Boccignone, Vittorio Caggiano, Carmine Cesarano, Vincenzo Moscato, Lucio Sansone
2005 Lecture Notes in Computer Science  
In this manner our tree provides a simple and practical solution to index clustered data and support efficient retrieval of the nearest neighbors in high dimensional object spaces.  ...  In this paper we introduce a new indexing approach to representing multimedia object classes generated by the Expectation Maximization clustering algorithm in a balanced and dynamic tree structure.  ...  An efficient index for a large data set, where data are described in high dimensional feature space, should allow to prune comparisons between data during the similarity search process by taking advantage  ... 
doi:10.1007/11551898_8 fatcat:z3zczsle6jeqbalg6l2lgvlkla

Measuring the Difficulty of Distance-Based Indexing [chapter]

Matthew Skala
2005 Lecture Notes in Computer Science  
Data structures for similarity search are commonly evaluated on data in vector spaces, but distance-based data structures are also applicable to non-vector spaces with no natural concept of dimensionality  ...  of index data structures on non-vector spaces by relating them to equivalent vector spaces.  ...  Chávez and Navarro prove bounds on the performance of several kinds of distance-based index structures for metric spaces in terms of ρ.  ... 
doi:10.1007/11575832_12 fatcat:xsbtpftj5zbdva7m7mvmof6eve

High-Dimensional Simplexes for Supermetric Search [chapter]

Richard Connor, Lucia Vadicamo, Fausto Rabitti
2017 Lecture Notes in Computer Science  
The n-point property is a generalisation of this where, for any (n + 1) objects in the space, there exists an n-dimensional simplex whose edge lengths correspond to the distances among the objects.  ...  3 In a metric space, triangle inequality implies that, for any three objects, a triangle with edge lengths corresponding to their pairwise distances can be formed.  ...  Acknowledgements The work was partially funded by Smart News, "Social sensing for breaking news", co-funded by the Tuscany region under the FAR-FAS 2014 program, CUP CIPE D58C15000270008.  ... 
doi:10.1007/978-3-319-68474-1_7 fatcat:tgzke3qwfbde5mhc26bww6rjjm

On the Necessary and Sufficient Conditions of a Meaningful Distance Function for High Dimensional Data Space [chapter]

Chih-Ming Hsu, Ming-Syan Chen
2006 Proceedings of the 2006 SIAM International Conference on Data Mining  
high dimensional space even with the commonly used Lp metric on the Euclidean space.  ...  It is empirically shown that the SDP significantly outperforms prior measures for its being stable in high dimensional data space and robust to noise, and is thus deemed more suitable for distance-based  ...  ., dimensionality resistant) distance function in high dimensional space.  ... 
doi:10.1137/1.9781611972764.2 dblp:conf/sdm/HsuC06 fatcat:tyantygzyjgqlevwwwwq4yzw7q

LazyLSH

Yuxin Zheng, Qi Guo, Anthony K.H. Tung, Sai Wu
2016 Proceedings of the 2016 International Conference on Management of Data - SIGMOD '16  
Different from previous LSH approaches which need to build one dedicated index for every query space, LazyLSH uses a single base index to support the computations in multiple p spaces, significantly reducing  ...  Current LSH-based approaches target at the 1 and 2 spaces, while as shown in previous work, the fractional distance metrics ( p metrics with 0 < p < 1) can provide more insightful results than the usual  ...  As the 1 metric is closer to the fractional distance metrics, in LazyLSH, we materialize the LSH index in the 1 space as our base index.  ... 
doi:10.1145/2882903.2882930 dblp:conf/sigmod/ZhengGTW16 fatcat:l5eispcnzvfkllq4l2jkygaqaq

Practical Construction of k-Nearest Neighbor Graphs in Metric Spaces [chapter]

Rodrigo Paredes, Edgar Chávez, Karina Figueroa, Gonzalo Navarro
2006 Lecture Notes in Computer Science  
Experiments suggest that it yields costs of the form c1n 1.27 distance computations for low and medium dimensional spaces, and c2n 1.90 for high dimensional ones. ⋆  ...  We present a general methodology to construct knngs that exploits several features of metric spaces.  ...  Acknowledgement We wish to thank Georges Dupret and Marco Patella for their valuable comments.  ... 
doi:10.1007/11764298_8 fatcat:c3lr4vzyxbae3ateuvssztbrvm
« Previous Showing results 1 — 15 out of 140,663 results