37,156 Hits in 6.3 sec

A kernel-based approach to document retrieval

Albert Gordo, Jaume Gibert, Ernest Valveny, Marçal Rusiñol
2010 Proceedings of the 8th IAPR International Workshop on Document Analysis Systems - DAS '10  
We show how our method based on similarity kernels outperforms the usual distance-based retrieval.  ...  In this paper we tackle the problem of document image retrieval by combining a similarity measure between documents and the probability that a given document belongs to a certain class.  ...  BACKGROUND ON KERNELS In this paper we adopt a kernel-based approach for document retrieval.  ... 
doi:10.1145/1815330.1815379 dblp:conf/das/GordoGVR10 fatcat:gxleqdnwibbrhihnpozsxavewm

Position-based contextualization for passage retrieval

David Carmel, Anna Shtok, Oren Kurland
2013 Proceedings of the 22nd ACM international conference on Conference on information & knowledge management - CIKM '13  
The core principle is to let any occurrence of a query term in a document affect the passage retrieval score, whether the occurrence is in the passage or not.  ...  We present a novel contextualization approach for passage retrieval.  ...  A novel contextualization approach We consider a novel contextualization approach for passage retrieval that is based on leveraging a fundamental principle underlying the locality-based similarity [8]  ... 
doi:10.1145/2505515.2507865 dblp:conf/cikm/CarmelSK13 fatcat:4sfwqafmp5buljhqeylgybhrxq

Proximity-based opinion retrieval

Shima Gerani, Mark James Carman, Fabio Crestani
2010 Proceeding of the 33rd international ACM SIGIR conference on Research and development in information retrieval - SIGIR '10  
We propose a proximity-based opinion propagation method to calculate the opinion density at each point in a document.  ...  In this paper we propose a simple probabilistic model for assigning relevant opinion scores to documents.  ...  This research was partly funded by the "Secrétariat d'étatà l'Éducation età la Recherche (SER)" and COST Action IC0702 "Combining Soft Computing Techniques and Statistical Methods to Improve Data Analysis  ... 
doi:10.1145/1835449.1835517 dblp:conf/sigir/GeraniCC10 fatcat:3weyimudl5dyjb4pgeau6g6i44

Combining visual features and text data for medical image retrieval using latent semantic kernels

Juan C. Caicedo, Jose G. Moreno, Edwin A. Niño, Fabio A. González
2010 Proceedings of the international conference on Multimedia information retrieval - MIR '10  
Then, a system to search using the query-by-example paradigm is evaluated instead of a keyword-based search.  ...  In this paper we propose an strategy to fuse visual features and unstructured-text data in a medical image retrieval system.  ...  This novel approach is based on a kernel method solution that allows to model complex document representations by operating with appropriate similarity measures.  ... 
doi:10.1145/1743384.1743442 dblp:conf/mir/CaicedoMNG10 fatcat:jvbk72tepvczvmsjjdizqobzje

One Class Classification Methods Based Non-Relevance Feedback Document Retrieval

Takashi Onoda, Hiroshi Murata, Seiji Yamada
2006 2006 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology Workshops  
However, the initial retrieved documents, which are displayed to a user, sometimes don't include relevant documents.  ...  In order to solve this problem, we propose a new feedback method using information of non-relevant documents only. We named this method nonrelevance feedback document retrieval.  ...  In general, since a user hardly describes a precise query in the first trial, interactive approach to modify the query vector by evaluation of the user on documents in a list of retrieved documents.  ... 
doi:10.1109/wi-iatw.2006.98 dblp:conf/iat/OnodaMY06a fatcat:xswgpadpdbhh5gzdhyntymmrna

Non-Relevance Feedback Document Retrieval based on One Class SVM and SVDD

T. Onoda, H. Murata, S. Yamada
2006 The 2006 IEEE International Joint Conference on Neural Network Proceedings  
However, the initial retrieved documents, which are displayed to a user, sometimes don't include relevant documents.  ...  Our proposed approach has been very useful for document retrieval with relevance feedback experimentally.  ...  In our experiments, One Class SVM based non-relevance feedback approach makes better performance than SVDD based approach for an interactive document retrieval.  ... 
doi:10.1109/ijcnn.2006.246829 dblp:conf/ijcnn/OnodaMY06 fatcat:22vxy47cejf5beksniudq5bnre

Fisher kernel based relevance feedback for multimodal video retrieval

Ionut Mironica, Bogdan Ionescu, Jasper Uijlings, Nicu Sebe
2013 Proceedings of the 3rd ACM conference on International conference on multimedia retrieval - ICMR '13  
This paper proposes a novel approach to relevance feedback based on the Fisher Kernel representation in the context of multimodal video retrieval.  ...  Hence during relevance feedback we create a new Fisher Kernel representation based on the most relevant examples.  ...  In this paper, we propose a new RF approach for video genre retrieval, using a combination of Fisher Kernels with SVM Classifiers.  ... 
doi:10.1145/2461466.2461478 dblp:conf/mir/MironicaIUS13 fatcat:ijtevl3zbre2xj2dmjqn5ozavq

Retrieving Passages and Finding Answers

Mostafa Keikha, Jae Hyun Park, W. Bruce Croft, Mark Sanderson
2014 Proceedings of the 2014 Australasian Document Computing Symposium on - ADCS '14  
Retrieving topically-relevant text passages in documents has been studied many times, but finding non-factoid, multiple sentence answers to web queries is a different task that is becoming increasingly  ...  As the first stage of developing retrieval models for "answer passages", we describe the process of creating a test collection of questions and multiple-sentence answers based on the TREC GOV2 queries  ...  We employ a homogeneity measure based on the length of the document to assign weights to each component.  ... 
doi:10.1145/2682862.2682877 dblp:conf/adcs/KeikhaPCS14 fatcat:ctp3zimpdbe4db4t23zoyh65ny

Conformer-Kernel with Query Term Independence for Document Retrieval [article]

Bhaskar Mitra, Sebastian Hofstatter, Hamed Zamani, Nick Craswell
2020 arXiv   pre-print
to BERT-based ranking models.  ...  We show that the Conformer's GPU memory requirement scales linearly with input sequence length, making it a more viable option when ranking long documents.  ...  This approach can not only negatively impact retrieval quality, but has been shown to specifically under-retrieve longer documents [Hofstätter et al., 2020a] .  ... 
arXiv:2007.10434v1 fatcat:vxpfn3fnwbge3fx5kiejezvhj4

Header-Words Based for Printed Arabic Document Images Retrieval System

2017 Iraqi Journal of Science  
In this paper, a printed Arabic document images retrieval system based on spotting the header words of official Arabic documents is proposed.  ...  Printed Arabic document image retrieval is a very important and needed system for many companies, governments and various users.  ...  The proposed system implement a new approach to retrieve the desired documents by extracting significant features from the header-words of document images.  ... 
doi:10.24996/ijs.2017.58.3c.18 fatcat:ytubhdaqynhjbdcmrbviirv72m

Combining content and structure similarity for XML document classification using composite SVM kernels

Saptarshi Ghosh, Pabitra Mitra
2008 Pattern Recognition (ICPR), Proceedings of the International Conference on  
Combination of structure and content features is necessary for effective retrieval and classification of XML documents.  ...  Classification experiments performed on the INEX 1.3 XML corpus, demonstrate that the composite kernel classifier achieves significantly better performance as compared to complex and time consuming approaches  ...  Other approaches include rule-based approaches like XRules [9] .  ... 
doi:10.1109/icpr.2008.4761539 dblp:conf/icpr/GhoshM08 fatcat:5ltulzed2naobjyaucdp2pn2k4

Document-Document similarity matrix and Multiple-Kernel Fuzzy C-Means Algorithm-based web document clustering for information retrieval
IJARCCE - Computer and Communication Engineering

2014 IJARCCE  
In this work, Document-Document similarity matrix and Multiple-Kernel Fuzzy C-Means Algorithm-based web document clustering is developed for information retrieval.  ...  Due to continuous development of World Wide Web, web database are growing massively where automatic grouping of web documents pose a new challenge for researchers to easily retrieve the information.  ...  DOCUMENT-DOCUMENT SIMILARITY MATRIX AND MULTIPLE-KERNEL FUZZY C-MEANS ALGORITHM TO WEB DOCUMENT CLUSTERING FOR INFORMATION RETRIEVAL This section presents the proposed document clustering approach using  ... 
doi:10.17148/ijarcce.2014.31054 fatcat:kxkj4xi2e5gktgdmtntumya7vu

An One Class Classification Approach to Non-relevance Feedback Document Retrieval [chapter]

Takashi Onoda, Hiroshi Murata, Seiji Yamada
2005 Lecture Notes in Computer Science  
The non-relevance feedback document retrieval is based on One-class Support Vector Machine.  ...  However, the initial retrieved documents, which are displayed to a user, sometimes don't include relevant documents.  ...  In general, since a user hardly describes a precise query in the first trial, interactive approach to modify the query vector using evaluation of the documents on a list of retrieved documents by a user  ... 
doi:10.1007/11540007_161 fatcat:vmiuefdzwbaddikcjl6ixfzlsy

Learning Similarity with Probabilistic Latent Semantic Analysis for Image Retrieval

2015 KSII Transactions on Internet and Information Systems  
It first derives Fisher kernel, a function over the parameters and variables, based on PLSA.  ...  Content based image retrieval (CBIR) is the most promising way to tackle this problem, where the most important topic is to measure the similarity of images so as to cover the variance of shape, color,  ...  Learning Fisher Kernel with PLSA In this section, we will proceed to derive the Fisher kernel based on PLSA and propose a supervised learning approach for the derived kernel.  ... 
doi:10.3837/tiis.2015.04.009 fatcat:vc3po2xtqrdbbdom4mjb5xkf64


Jiashu Zhao, Jimmy Xiangji Huang, Ben He
2011 Proceedings of the 34th international ACM SIGIR conference on Research and development in Information - SIGIR '11  
Term proximity retrieval rewards a document where the matched query terms occur close to each other.  ...  We propose a Cross Term Retrieval (CRTER) model that combines the Cross Terms' information with basic probabilistic weighting models to rank the retrieved documents.  ...  Therefore, traditional proximity-based approaches may not be able to boost the relevant documents of this specific topic.  ... 
doi:10.1145/2009916.2009941 dblp:conf/sigir/ZhaoHH11 fatcat:oczgvpy3wrbdth7aeap6ml4vre
« Previous Showing results 1 — 15 out of 37,156 results