15 Hits in 7.3 sec

Using Ontology Dimensions and Negative Expansion to solve Precise Queries in the ImageCLEF Medical Task

Jean-Pierre Chevallet, Joo-Hwee Lim, Saïd Radhouani
2005 Conference and Labs of the Evaluation Forum  
Our hypothesis is that precise document should emphasis only few terms of a given dimension. To compute this new constraint, we have set up a negative weight query expansion.  ...  To our opinion, it shows that for a given domain, adding explicit knowledge stored into an ontology tree, enable to classify the importance of terms used in the query and enhance the finale average precision  ...  The increase of average precision using negative query expansion with an ontology is then encouraging.  ... 
dblp:conf/clef/ChevalletLR05a fatcat:f7ildx7gs5hddf6jzkqjcqhfja

Combining Textual and Visual Ontologies to Solve Medical Multimodal Queries

Said Radhouani, Joo HweeLim, Jean-pierre Chevallet, Gilles Falquet
2006 2006 IEEE International Conference on Multimedia and Expo  
In order to solve medical multimodal queries, we propose to split the queries in different dimensions using ontology.  ...  We extract both textual and visual terms depending on the ontology dimension they belong to. Based on these terms, we build different sub queries each corresponds to one query dimension.  ...  We also present how the fusion of ranked lists, generated using both text and image indexes, can contribute to solve precise multimodal queries and improve the retrieval performance.  ... 
doi:10.1109/icme.2006.262915 dblp:conf/icmcs/RadhouaniLCF06 fatcat:ccg2hbqflnbadlxwocib6haytq

Using heterogeneous annotation and visual information for the benchmarking of image retrieval systems

Henning Müller, Paul Clough, William Hersh, Thomas Deselaers, Thomas M. Lehmann, Bruno Janvier, Antoine Geissbuhler, Simone Santini, Raimondo Schettini, Theo Gevers
2006 Internet Imaging VII  
This, in combination with high-level semantic queries, makes visual/textual combinations almost useless as the information need can often be solved using just textual features.  ...  Two collections have an annotation per case and not per image, which is normal in the medical domain, making it difficult to relate parts of the accompanying text to corresponding images.  ...  and 205321-109304/1.  ... 
doi:10.1117/12.660259 fatcat:6yrpnkatabe5hfi54kmnwajofy

Overview of the ImageCLEF 2012 Flickr Photo Annotation and Retrieval Task

Bart Thomee, Adrian Popescu
2012 Conference and Labs of the Evaluation Forum  
ImageCLEF's Flickr Photo Annotation and Retrieval task aims to advance the state of the art in multimedia research by providing a challenging benchmark for visual concept detection, annotation and retrieval  ...  image annotation, while the objective of second subtask was to correctly retrieve relevant images for conceptoriented queries inspired by what people actually search for on the internet.  ...  Finally, we are very grateful to the many crowdsource workers who performed the actual relevance assessments, as well as to the Flickr users whose photos we used in this task.  ... 
dblp:conf/clef/ThomeeP12 fatcat:cd2i3icgfnfhtawvwwgsake4hy

A graph-based approach for the retrieval of multi-modality medical images

Ashnil Kumar, Jinman Kim, Lingfeng Wen, Michael Fulham, Dagan Feng
2014 Medical Image Analysis  
The extraction and visualisation of knowledge from these repositories is a major challenge in the modern, digital world. In the medical domain, im-"It's bigger on the inside!"  ...  Without such stalwart friends and colleagues, I would have been left beleaguered by the roadside and it is thanks to them that I stand near the end of my journey.  ...  Experiments on the ImageCLEF medical data set showed that pre-filtering resulted in higher precision and recall than executing queries on the entire data set.  ... 
doi:10.1016/ pmid:24378541 fatcat:mqzxyjza4bhwdk7iat5veyaqfq

From CLEF to TrebleCLEF: the Evolution of the Cross-Language Evaluation Forum

Nicola Ferro, Carol Peters
2008 NTCIR Conference on Evaluation of Information Access Technologies  
In the first part of the paper, we provide a brief overview of the entire activity and summarise the main achievements; in the second part, we focus our attention on the Ad Hoc track with the aim of showing  ...  how the results of evaluation can be exploited to increase understanding of the many issues involved in multilingual retrieval system development.  ...  We should like to express out gratitude to all those who have been involved in the coordination of this activity over the years.  ... 
dblp:conf/ntcir/FerroP08 fatcat:h3p27ct6grgdnoe4sr52ce77va

A Survey of Quantum Theory Inspired Approaches to Information Retrieval [article]

Sagar Uprety and Dimitris Gkoumas and Dawei Song
2020 arXiv   pre-print
This paper presents a survey of the research done in this area, aiming to show the landscape of the field and draw a road-map of future directions.  ...  Since 2004, researchers have been using the mathematical framework of Quantum Theory (QT) in Information Retrieval (IR). QT offers a generalized probability and logic framework.  ...  The authors are hopeful that this literature survey is able to provide a clear picture of the quantum-inspired IR field and set a road-map for researchers to take this field forward.  ... 
arXiv:2007.04357v1 fatcat:uc6mi6mlizfivk2v3reb474s4q

Gravitate Project D3.1 Report On Shape Analysis And Matching And On Semantic Matching

Silvia Biasotti, Andrea Cerri, Chiara Catalano E., Bianca Falcidieno, Maria Laura Torrente, Stuart Middleton E., Leo Dorst, Ilan Shimshoni, Ayellet Tal, Dominic Oldman
2016 Zenodo  
These fields are relatively disjoint, and the research and development challenge of GRAVITATE is precisely to merge them.  ...  This state of the art report describes the techniques of shape analysis, and of metadata search that have been already implemented in cultural heritage or we think are useful for the GRAVITATE project.  ...  These fields are relatively disjoint at the moment, and the research and development challenge of GRAVITATE is precisely to merge them for our chosen tasks.  ... 
doi:10.5281/zenodo.55432 fatcat:tc3rsykiufezxeper7ldew2ibu

Exploiting Social Networks. Technological Trends (Habilitation Thesis) [article]

Adrian Iftene
2020 arXiv   pre-print
The work was validated with good results in evaluation campaigns like CLEF (Question Answering, Image CLEF, LifeCLEF, etc.), SemEval (Sentiment and Emotion in text, Anorexia, etc.).  ...  The habilitation thesis presents two main directions: 1.  ...  Acknowledgments To my collaborators and to my family  ... 
arXiv:2004.14386v1 fatcat:faoyxt42e5b5no57v6r5arnu5a

Semantics and statistics for automated image annotation

Ainhoa Llorente Coto
semantics of the words used in the annotation process, is able to improve the performance of probabilistic annotation systems.  ...  In particular, approaches based on Bayesian theory use machine-learning techniques to learn statistical models from a training set of pre-annotated images and apply them to generate annotations for unseen  ...  Medical Retrieval Task The medical retrieval task of ImageCLEF 2010 will use a similar database to 2008 and 2009 but with a larger number of images.  ... 
doi:10.21954/ fatcat:xtz4bm6wqjgs5m4yvffj6th62m

Machine learning methods for multimedia information retrieval [article]

Bálint Zoltán Daróczy
2017 pre-print
but not limited to the ones presented in this thesis.  ...  Since the used similarity graphs (Section 4.2) are greatly constrained for computational purposes, we would like to continue work with more complex, evolving and capable graphs and apply for different  ...  query expansion formula of [Xu and Croft, 1996] .  ... 
doi:10.15476/elte.2016.086 arXiv:1705.04964v1 fatcat:ccvjysaxpjfbbkq3di4qmtwqpy

Multimodal News Summarization, Tracking and Annotation Incorporating Tensor Analysis of Memes

Chun-Yu Tsai
We show the method is fast, can be tuned to give preferences to any subset of its four dimensions, and exceeds three existing methods in performance.  ...  The dimensions correspond to extracted visual memes, verbal tags, time periods, and cultures.  ...  On the image side, the output dimensions of the two hidden layers we use are [2048, 512]; on the tag side, Figure 6 . 3 : 63 We use an image in culture 1 (say, U.S.) to query texts in culture 2 (say,  ... 
doi:10.7916/d8ff44n7 fatcat:4lr45mn3wbhspmyit3kkvujlz4

Exploiting Social Networks for Recommendation in Online Image Sharing Systems

Adam Rae
The online photo sharing site Flickr provides rich and varied data that can be used by researchers to analyse and understand users' interactions with images and with each other.  ...  I analyse such data by modelling the connections between users as multigraphs and exploiting the resultant topologies to produce features that can be used to train recommender systems based on machine  ...  I realised I had been on a trajectory throughout my education, aiming for this particular milestone and that it felt natural to pursue what would (I hoped) be a rewarding experience.  ... 
doi:10.21954/ fatcat:36i6yfiyebaild242cyajki2hy

Dagstuhl Reports, Volume 7, Issue 7, July 2017, Complete Issue [article]

We have demonstrated two tasks: 1) Given a query image (without any accompanying text), we retrieve textual descriptions that correspond to the visual attributes in the visual query; and 2) given a textual  ...  Another well-known example is the ImageCLEF 2009 Photo Task, which revolved around image diversification [23] .  ...  In JATE2, instead of using the annotation file "GENIAcorpus302.xml", the 'concept.txt' containing a breakdown list of GENIA concepts and relations (more like ontology) are used as the "Gold Standard" (  ... 
doi:10.4230/dagrep.7.7 fatcat:ve4n3wvvk5bkfgae36nsto7sre

OASIcs-OpenAccess Series in Informatics

Andrew Jones, Andrew Jones, Daniel Cremers, Barbara Hammer, Marc Langheinrich, Dorothea Wagner
1998 Conference Proceedings   unpublished
Acknowledgements The author would like to particularly thank his primary supervisor, Dr.  ...  Acknowledgements We would like to thank Domenico Corapi for his help and discussion on TAL and ASPAL.  ...  The main purpose of this global ontology is to provide a unified query interface for the local ontologies.  ...