Filters








134,232 Hits in 4.5 sec

Multimedia Indexing, Search, and Retrieval in Large Databases of Social Networks [chapter]

Theodoros Semertzidis, Dimitrios Rafailidis, Eleftherios Tiakas, Michael G. Strintzis, Petros Daras
2012 Computer Communications and Networks  
In the first part pure multimedia content retrieval issues are presented, while in the second part, the social aspects and new, interesting views on multimedia retrieval in the large social media databases  ...  Note that the comparison with M-Tree is omitted, since M-Tree belongs to the family of exact similarity search and its retrieval accuracy is always equal to 100%.  ...  Additionally, the M-Tree has been further extended, in order to support both exact and approximate strategies, while preserving the same indexing structure of the M-Tree file.  ... 
doi:10.1007/978-1-4471-4555-4_3 dblp:series/ccn/SemertzidisRTSD13 fatcat:6l3whv5qcjgshmjwcqh6dgelb4

Adding Flexibility to Structure Similarity Queries on XML Data [chapter]

Paolo Ciaccia, Wilma Penzo
2002 Lecture Notes in Computer Science  
degree of correctness and completeness of results with respect to a given query, as well as the degree of cohesion of data retrieved.  ...  In this paper we present an approach that faces this issue in a twofold fashion: 1) it exploits new approximations on data structure; 2) it provides a relevance ranking method that takes into account the  ...  Approximate Embedding of Query Trees The similarity function we propose for retrieval and scoring of embeddings of a query tree in a document tree, is given by the SAT ES (Scored Approximate Tree Embedding  ... 
doi:10.1007/3-540-36109-x_10 fatcat:bljggljjhjgltkaa4rbauavdge

Error-tolerant retrieval of trees

K. Oflazer
1997 IEEE Transactions on Pattern Analysis and Machine Intelligence  
Index Terms-Example-based machine translation, approximate tree comparison, retrieval from lexical databases, tree databases.  ...  For large randomly generated synthetic tree databases (some having tens of thousands of trees), and on databases constructed from Wall Street Journal treebank, it can retrieve for trees with a small error  ...  a query tree exactly but rather the retrieval of trees that match approximately (with the 1.  ... 
doi:10.1109/34.643897 fatcat:kckn7mhz7jgzzeifthx2aruo2u

Improving retrieval accuracy of Hierarchical Cellular Trees for generic metric spaces

Carles Ventura, Verónica Vilaplana, Xavier Giró-i-Nieto, Ferran Marqués
2013 Multimedia tools and applications  
Therefore, MAMs are specially useful for Content-Based Image Retrieval systems based on features which use non L p norms as similarity measures.  ...  retrieval time and the retrieval accuracy.  ...  In contrast with HCT, M-tree uses the Most Similar Nucleus (MS-Nucleus) search algorithm, which consists on descending only through the HCT branch which gives the most similar element at each level.  ... 
doi:10.1007/s11042-013-1686-3 fatcat:pfybxs2phrhzhhy3jrrkunrira

Unified framework for fast exact and approximate search in dissimilarity spaces

Tomáš Skopal
2007 ACM Transactions on Database Systems  
), e.g. the M-tree.  ...  To date, the mentioned aspects of similarity retrieval have been solved separately, i.e. exact vs. approximate search or metric vs. non-metric search.  ...  respect to the effectiveness and efficiency of retrieval by two MAMs (the M-tree and PMtree).  ... 
doi:10.1145/1292609.1292619 fatcat:nenccwjhvjgnlcdbz4otf3pzze

A tree algorithm for nearest neighbor searching in document retrieval systems

Caroline M. Eastman, Stephen F. Weiss
1978 Proceedings of the 1st annual international ACM SIGIR conference on Information storage and retrieval - SIGIR '78  
A nearest neighbors associative retrieval algorithm, suitable for document retrieval using similarity matching, is described.  ...  A series of experiments with a small collection confirm the predictions made using the analytic model; k is approximately 4 in this situation.  ...  The number of nearest neighbors actually retrieved was less than m if there were fewer than m documents with similarity greater than 0 to the query.  ... 
doi:10.1145/800096.803139 dblp:conf/sigir/EastmanW78 fatcat:s6z5axdzezb6xbty3tadynzc4u

A tree algorithm for nearest neighbor searching in document retrieval systems

Caroline M. Eastman, Stephen F. Weiss
1978 SIGIR Forum  
A nearest neighbors associative retrieval algorithm, suitable for document retrieval using similarity matching, is described.  ...  A series of experiments with a small collection confirm the predictions made using the analytic model; k is approximately 4 in this situation.  ...  The number of nearest neighbors actually retrieved was less than m if there were fewer than m documents with similarity greater than 0 to the query.  ... 
doi:10.1145/1013234.803139 fatcat:jgtydnj435ci5fcsvmsn5ynfka

Semantic retrieval of multimedia data

Samira Hammiche, Salima Benbernou, Mohand-Sa�d Hacid, Athena Vakali
2004 Proceedings of the 2nd ACM international workshop on Multimedia databases - MMDB '04  
We propose a framework for querying multimedia data based on a tree embedding approximation algorithm, combining the MPEG-7 standard and an ontology.  ...  This paper deals with the problem of finding multimedia data that fulfill the requirements of user queries. We assume both the user query and the multimedia data are expressed by MPEG-7 standard.  ...  A retrieval process should retrieve not only the results that match exactly the query, but also answers with a structure and content similar to the conditions specified by the query.  ... 
doi:10.1145/1032604.1032612 dblp:conf/mmdb/HammicheBHV04 fatcat:ryeivaocdnexhiglk2ibwxarfa

Anytime K-Nearest Neighbor Search for Database Applications

Weijia Xu, Daniel P. Miranker, Rui Mao, Smriti Ramakrishnan
2008 First International Workshop on Similarity Search and Applications (sisap 2008)  
In support of these applications, we explored a search strategy for knearest neighbor searches with MVP-trees that greedily identifies k answers and then improves the answer set monotonically.  ...  Many contemporary database applications require similarity-based retrieval of complex objects where the only usable knowledge of its domain is determined by a metric distance function.  ...  By the construction of an MVP-tree: min(i,j) d(m, p i ) max(i,j) .  ... 
doi:10.1109/sisap.2008.11 dblp:conf/sisap/XuMMR08 fatcat:nxr4l7iv7bbupgj5gw6olk6xyu

Forest hashing: Expediting large scale image retrieval

Jonathan Springer, Xin Xin, Zhu Li, Jeremy Watt, Aggelos Katsaggelos
2013 2013 IEEE International Conference on Acoustics, Speech and Signal Processing  
This combination gives us the local approximate nearest neighbor accuracy of kd-trees with the computational efficiency of hashing techniques.  ...  Index Terms-image retrieval, kd-tree, spectral hashing, forest hashing.  ...  We believe that building trees with different subsets of dimensions and utilizing these trees to find similar descriptors will give a higher number of accurate results in retrieval and will avoid a common  ... 
doi:10.1109/icassp.2013.6637938 dblp:conf/icassp/SpringerXLWK13 fatcat:pj3eixi27rcnjep7er2tr3hrbe

Lucene for Approximate Nearest-Neighbors Search on Arbitrary Dense Vectors [article]

Tommaso Teofili, Jimmy Lin
2019 arXiv   pre-print
vectors, and k-d trees coupled with dimensionality reduction.  ...  We demonstrate three approaches for adapting the open-source Lucene search library to perform approximate nearest-neighbor search on arbitrary dense vectors, using similarity search on word embeddings  ...  For a given query vector q, the goal is to retrieve its top k most similar vectors (in terms of cosine similarity), where we determine the "ground truth" by brute force (in the case of k-d trees, on the  ... 
arXiv:1910.10208v2 fatcat:wuuh5do64nakhoumfzsf5xkebe

Anytime k-nearest neighbor search for database applications

Weijia Xu, Daniel Miranker, Rui Mao, Smriti Ramakrishnan
2008 2008 IEEE 24th International Conference on Data Engineering Workshop  
In support of these applications, we explored a search strategy for knearest neighbor searches with MVP-trees that greedily identifies k answers and then improves the answer set monotonically.  ...  Many contemporary database applications require similarity-based retrieval of complex objects where the only usable knowledge of its domain is determined by a metric distance function.  ...  By the construction of an MVP-tree: min(i,j) d(m, p i ) max(i,j) .  ... 
doi:10.1109/icdew.2008.4498354 dblp:conf/icde/XuMMR08 fatcat:vppvtvfoinghre3nfmx6phqyjq

Video Data Indexing by 2D C-Trees

FANG-JUNG HSU, SUH-YIN LEE, BAO-SHUH LIN
1998 Journal of Visual Languages and Computing  
The similarity retrieval of video matching problem becomes the problem of video sequence matching by computing the similarity, or the minimum cost of matched frames.  ...  The matching schemes can be modified and extended to approximate sequence matching by computing the partial distance for providing a comprehensive retrieval of video data.  ...  The approximate sequence matching provides a comprehensive operation for retrieving video data.  ... 
doi:10.1006/jvlc.1998.0089 fatcat:klibq5uxr5btnimxs3nlosxwv4

Multimodal similarity-preserving hashing [article]

Jonathan Masci and Michael M. Bronstein and Alexander A. Bronstein and Jürgen Schmidhuber
2012 arXiv   pre-print
We show experimentally that our method significantly outperforms state-of-the-art hashing approaches on multimedia retrieval tasks.  ...  The proposed approach is based on a novel coupled siamese neural network architecture and allows unified treatment of intra- and inter-modality similarity learning.  ...  The settings with α X = α Y = m X = m Y = 0 correspond to CM-NN.  ... 
arXiv:1207.1522v1 fatcat:6kd7jaaoqvgwhabrkqdqawsnnq

Efficient retrieval of recommendations in a matrix factorization framework

Noam Koenigstein, Parikshit Ram, Yuval Shavitt
2012 Proceedings of the 21st ACM international conference on Information and knowledge management - CIKM '12  
To this end, we propose two techniques for efficient search -(i) We index the item vectors in a binary spatial-partitioning metric tree and use a simple branchand-bound algorithm with a novel bounding  ...  We obtain a theoretical error bound which determines the quality of any approximate result and use it to control the approximation.  ...  In addition, we define a secondary metric (M edianRank) which denotes the preference of the approximated recommendations with respect to the rest of the items: M edianRank(K) median {∪uRank(Lrec(u))}  ... 
doi:10.1145/2396761.2396831 dblp:conf/cikm/KoenigsteinRS12 fatcat:a6ppf7syazfsfhbqr4ictkm4va
« Previous Showing results 1 — 15 out of 134,232 results