Filters








86,911 Hits in 5.6 sec

Re-ranking search results using an additional retrieved list

Lior Meister, Oren Kurland, Inna Gelfer Kalmanovich
2010 Information retrieval (Boston)  
We present a novel approach to re-ranking a list that was retrieved in response to a query so as to improve precision at the very top ranks.  ...  In contrast to commonly-used methods for fusion of retrieved lists, our approach also exploits inter-document-similarities between the lists -a potentially rich source of additional information.  ...  We present a novel approach to re-ranking an initially retrieved list.  ... 
doi:10.1007/s10791-010-9150-8 fatcat:f3tpdowtofge7cnwtcuvvh4nsa

Aggregation of Multiple Pseudo-Relevance Feedbacks for Image Search Reranking

Wei-Chao Lin
2019 IEEE Access  
Since there are several different strategies for performing PRF leading to different search results, in this paper we focus on image search re-ranking by search result aggregation as a hybrid approach.  ...  INDEX TERMS Image retrieval, re-ranking, pseudo relevance feedback, Borda count.  ...  In particular, the top-k pseudo positive feedback set is used alone, resulting in a new ranked retrieval list (i.e. retrieval list 2).  ... 
doi:10.1109/access.2019.2942142 fatcat:jqucuyx4vrcqpg7jzv7vyv3ndu

The search duel

Peter Izsak, Fiana Raiber, Oren Kurland, Moshe Tennenholtz
2014 Proceedings of the 37th international ACM SIGIR conference on Research & development in information retrieval - SIGIR '14  
That is, the goal is to produce in response to a query a document result list whose effectiveness does not fall much behind that of the strong search engine; and, which is quite different than that of  ...  How can a search engine with a relatively weak relevance ranking function compete with a search engine that has a much stronger ranking function?  ...  overlap with the result list produced by the strong search engine, measured using OV@10. weak;rerank .  ... 
doi:10.1145/2600428.2609474 dblp:conf/sigir/IzsakRKT14 fatcat:mi5ei3ic5nfeffujxd6shd4uqu

Diversity, Assortment, Dissimilarity, Variety: A Study of Diversity Measures Using Low Level Features for Video Retrieval [chapter]

Martin Halvey, P. Punitha, David Hannah, Robert Villa, Frank Hopfgartner, Anuj Goyal, Joemon M. Jose
2009 Lecture Notes in Computer Science  
In this paper we present a number of methods for re-ranking video search results to introduce diversity into the set of search results.  ...  The responses from the users indicate that they find the more diverse search results extremely useful.  ...  C refers to the original result list, R is the re-ranked result list, z refers to an image in C. Equation 7 was used for HTD and HTRR. Equation 6 is the most different of the measures.  ... 
doi:10.1007/978-3-642-00958-7_14 fatcat:kkoye2avwrdvzmujovqtyfkt54

Entity-Based Stochastic Analysis Of Search Results For Query Expansion And Results Re-Ranking

Pavlos Fafalios, Yannis Tzitzikas
2015 Zenodo  
The result of this analysis is exploited in two different contexts: for automatic query expansion and for re-ranking a set of retrieved results.  ...  list of results by moving relevant but low-ranked hits in higher positions.  ...  Re-ranking the Retrieved Results. All document-nodes have received a final PageRank score. We exploit these scores for re-ranking the list of retrieved results.  ... 
doi:10.5281/zenodo.804259 fatcat:kzye5sm7i5crzcnc4i5y5lqpsy

DCU-TCD@LogCLEF 2010: Re-ranking Document Collections and Query Performance Estimation

Johannes Leveling, M. Rami Ghorab, Walid Magdy, Gareth J. F. Jones, Vincent Wade
2010 Conference and Labs of the Evaluation Forum  
The results of the collection re-ranking show a significant improvement in Mean Average Precision (MAP) over the original collection ranking of TEL.  ...  Moreover, the investigation involved assigning different weights to the three attributes in a scoring function that was used to re-rank the collections displayed to the user according to the language and  ...  Result Re-ranking Result re-ranking is one of the well-known techniques used for search personalisation. [12] [13] [14] .  ... 
dblp:conf/clef/LevelingGMJW10 fatcat:3xev2bhiuzbr3dxvug3uacttcu

Personalized Concept-Based Search and Exploration on the Web of Data Using Results Categorization [chapter]

Melike Sah, Vincent Wade
2013 Lecture Notes in Computer Science  
Within the selected concept lens; more relevant results are included using results re-ranking and query expansion, as well as relevant concept lenses are suggested to support results exploration.  ...  The results showed that the proposed approach performs significantly better than a non-adaptive baseline conceptbased search and traditional ranked list presentation.  ...  In addition, within the selected lens; the ranked list is personalized to push up the relevant results and the category label is used to generate an expanded query to retrieve more relevant results.  ... 
doi:10.1007/978-3-642-38288-8_36 fatcat:2kl6pruezrhmnplxx2qignmczy

Personalized Concept-Based Search on the Linked Open Data

Melike Sah, Vincent Wade
2016 Social Science Research Network  
Especially, the commonly used ranked lists presentation of current LOD search engines is not effective for search tasks in unfamiliar domains.  ...  In this paper, we evaluate the impact of personalized concept-based search in terms of task assistance and user satisfaction, while comparing with the ranked lists.  ...  This gain was mainly obtained by the personalization; results re-ranked as well as more relevant results were automatically pushed on top of the search list using category/results re-ranking and query  ... 
doi:10.2139/ssrn.3199219 fatcat:k3pt5v4wjjhpdeqsujgvo4uua4

York University at TREC 2006: Enterprise Email Discussion Search

Yu Fan, Xiangji Huang, Aijun An
2006 Text Retrieval Conference  
First, we make use of the thread structure in the emails to re-rank the documents retrieved by Okapi.  ...  the stop word list used.  ...  After Okapi outputs a ranked list for a query, we re-rank the documents in the list by using the thread structure. The re-ranking method is straightforward.  ... 
dblp:conf/trec/FanHA06 fatcat:vviulfo7iven7f4dehcf365xj4

Improving Retrieval Quality Using Pseudo Relevance Feedback in Content-Based Image Retrieval

Dinesha Chathurani Nanayakkara Wasam Uluwitige, Timothy Chappell, Shlomo Geva, Vinod Chandran
2016 Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval - SIGIR '16  
An original, simple yet effective rank-based PRF mechanism (RB-PRF) that takes into account the initial rank order of each image to improve retrieval accuracy is proposed.  ...  This RB-PRF mechanism innovates by making use of binary image signatures to improve retrieval precision by promoting images similar to highly ranked images and demoting images similar to lower ranked images  ...  This is then used to re-rank the initial search results.  ... 
doi:10.1145/2911451.2914747 dblp:conf/sigir/UluwitigeCGC16 fatcat:ohil4r6xnvcinn5uqdjzdiw42i

Improving Web Image Search Re-ranking

L Ramadevi
2013 IOSR Journal of Computer Engineering  
We propose an adaptive visual similarity to re-rank the text based search results.  ...  A query image is first categorized as one of several predefined intention categories, and a specific similarity measure has been used inside each of category to combine the image features for re-ranking  ...  Relevance Model Let us formulate the web image retrieval re-ranking problem in a more formal way.  ... 
doi:10.9790/0661-1314450 fatcat:cux3f3smy5bw3paiiayv3k4aje

Characterizing Search Behavior in Productivity Software

Horatiu Bota, Adam Fourney, Susan T. Dumais, Tomasz L. Religa, Robert Rounthwaite
2018 Proceedings of the 2018 Conference on Human Information Interaction&Retrieval - CHIIR '18  
they want to execute, retrieve help documentation or access web results in a unified interface.  ...  Our findings show that users engage primarily in command search, and that re-accessing commands through search is a frequent behavior.  ...  Selecting which result lists to re-rank is not trivial, given that more than 70% of searches retrieved optimal rankings (i.e. clicked command returned at top rank), and as such, re-ranking these searches  ... 
doi:10.1145/3176349.3176395 dblp:conf/chiir/BotaFDRR18 fatcat:hhpayyamn5bblekftbuhln5wvq

Multimedia search reranking

Tao Mei, Yong Rui, Shipeng Li, Qi Tian
2014 ACM Computing Surveys  
Discovering knowledge or visual patterns from such a noisy ranked list to guide the re-ranking process is difficult. Numerous techniques have been developed for visual search re-ranking.  ...  Such a problem is challenging because the initial search results often have a great deal of noise.  ...  In the most common formulation, the re-ranking problem can be reduced to a problem of re-estimating relevance for each document that has been ranked in an initial search result (e.g., a ranked list of  ... 
doi:10.1145/2536798 fatcat:6kmga3jo4fa2tp4354wukfuzja

Cross-Modal Person Search: A Coarse-to-Fine Framework using Bi-Directional Text-Image Matching

Xiaojing Yu, Tianlong Chen, Yang Yang, Michael Mugo, Zhangyang Wang
2019 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)  
To improve the accuracy off an image-based retrieval task, e.g., person re-identification (Person Re-Id), re-ranking is known to be an effective post-processing tool.  ...  In this paper, we extend re-ranking from uni-modal retrieval to cross-modal retrieval for the first time, and develop a bi-directional coarse-to-fine framework (BCF) for cross-modal person search.  ...  Among many successful practices, re-ranking methods are widely used to refine the ranking list in Person Re-Id and other image-to-image retrieval tasks.  ... 
doi:10.1109/iccvw.2019.00223 dblp:conf/iccvw/YuCYMW19 fatcat:tcizrmwbcvgnjbxnxe7l5zlh2y

USTB at Social Book Search 2016 Suggestion Task: Active Books Set and Re-ranking

Shao-Hui Feng, Bo-Wen Zhang, Xu-Cheng Yin, Zan-Xia Jin, Jian-Lin Jin, Jianwei Wu, Le-Le Zhang, Hao-Jie Pan, Fan Fang, Fang Zhou
2016 Conference and Labs of the Evaluation Forum  
The queries input to search engine were composed by key words from topics, and then we performed re-ranking models(popularity related) on Galago searching results on enriched XML index by 14 different  ...  We used Galago search from the index which was built with active books.  ...  When using the re-ranking approach, if a work of the result is contained in the topics list.  ... 
dblp:conf/clef/FengZYJJWZPFZ16 fatcat:y4e34hor2rbxhhprpnojjqj2s4
« Previous Showing results 1 — 15 out of 86,911 results