Blog, Enterprise, and Relevance Feedback
Published 1 January 2008
Krisztian Balog, Edgar Meij, Wouter Weerkamp, Jiyin He, Maarten de Rijke
Citations2
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
The main preliminary conclusions are that estimating mixture weights for external expansion in blog post retrieval is non-trivial and more analysis is needed to find out why it works better for blog distillation than for blog post re- trieval.
Abstract
We describe the participation of the University of Amsterdam's ILPS group in the blog, enterprise and relevance feedback track at TREC 2008. Our main preliminary conclusions are that estimating mixture weights for external expansion in blog post retrieval is non-trivial and we need more analysis to find out why it works better for blog distillation than for blog post re- trieval. For the relevance feedback track we ob- serve two things: (i) in terms of statMAP, a larger number of judged non-relevant documents im- proves retrieval effectiveness and (ii) on the TREC Terabyte topics, we can effectively replace the estimates on the judged non-relevant documents with estimations on the document collection. Fi- nally, since the enterprise track did not have any results yet, we only described our participation and do not draw any conclusions.
Keywords
Computer Science
ACM SIGIR ForumRelevance-Based Language Models
1,449 Citations2017Victor Lavrenko, W. Bruce Croft
This work proposes a novel technique for estimating a relevance model with no training data and demonstrates that it can produce highly accurate relevance models, addressing important notions of synonymy and polysemy.
Model-based feedback in the language modeling approach to information retrieval
799 Citations2001ChengXiang Zhai, John Lafferty
This paper proposes and evaluates two different approaches to updating a query language model based on feedback documents, one based on a generative probabilistic model of feedback documents and onebased on minimization of the KL-divergence over feedback documents.
ACM SIGIR ForumDocument Language Models, Query Models, and Risk Minimization for Information Retrieval
774 Citations2017John Lafferty, ChengXiang Zhai
A framework for information retrieval that combines document models and query models using a probabilistic ranking function based on Bayesian decision theory is presented and an operational retrieval model that extends recent developments in the language modeling approach to information retrieval is suggested.
Formal models for expert finding in enterprise corpora
552 Citations2006Krisztian Balog, Leif Azzopardi +1 more
This work presents two general strategies to expert searching given a document collection which are formalized using generative probabilistic models, and shows that the second strategy consistently outperforms the first.
A study of methods for negative relevance feedback
128 Citations2008Xuanhui Wang, Hui Fang +1 more
Experimental results on several TREC collections show that language model based negative feedback methods are generally more effective than those based on vector-space models, and using multiple negative models is an effective heuristic for negative feedback.
UvA-DARE (University of Amsterdam)Credibility Improves Topical Blog Post Retrieval
97 Citations2008Wouter Weerkamp, Maarten de Rijke
This work describes how to estimate textual credibility indicators and how to integrate them into a retrieval approach based on language models, and shows that both groups of credibility indicators significantly improve retrieval effectiveness.
ACM SIGIR ForumVariations on language modeling for information retrieval
79 Citations2005Wessel Kraaij
This dissertation makes a contribution to the field of language modeling (LM) for IR, which views both queries and documents as instances of a unigram language model and defines the matching function between a query and each document as the probability that the query terms are generated by the document language model.
Better than the real thing?
74 Citations2005Oren Kurland, Lillian Lee +1 more
A novel approach to pseudo-feedback-based ad hoc retrieval that uses language models induced from both documents and clusters, rivaling those of a number of previously-proposed algorithms, including the standard language-modeling approach.
People search in the enterprise
70 Citations2007Krisztian Balog
This thesis focuses on one particular type of entity: people, which brings new, exciting challenges to the fields of Information Retrieval and Information Extraction.
ACM Transactions on Information SystemsThe effect of accessing nonmatching documents on relevance feedback
54 Citations1997Mark Dunlop
The results shown here are encouraging, as they can be justified by an intuitive view of the process, but for negative feedback the results show behavior that cannot easily be justified and that varies greatly depending on the model of feedback used.
A few examples go a long way
45 Citations2008Krisztian Balog, Wouter Weerkamp +1 more
This work addresses a specific enterprise document search scenario, where the information need is expressed in an elaborate manner using a short query together with examples of key reference pages, and investigates how the examples can be utilized to improve the end-to-end performance on the document retrieval task.
Non-local evidence for expert finding
26 Citations2008Krisztian Balog, Maarten de Rijke
This work proposes modified expert retrieval models that are capable of incorporating both local and non-local evidence of expertise, i.e., evidence obtained using information not available in the immediate proximity of a candidate expert's name occurrence or even on the same page on which the name occurs.
Frontiers in artificial intelligence and applicationsFinding Key Bloggers, One Post At A Time
22 Citations2008Wouter Weerkamp, Krisztian Balog +1 more
A microelectronic package for the protection, housing, cooling and interconnection of a microelectronics chip.
Information Processing & ManagementRe-examining the effects of adding relevance information in a relevance feedback environment
21 Citations2008Weng-Kee Wong, Robert W. P. Luk +3 more
This paper presents an investigation about how to automatically formulate effective queries using full or partial relevance information in the context of relevance feedback (RF), and achieves a new level of retrieval effectiveness which is about 55-60% MAP instead of 40+% in the previous findings.
Data Archiving and Networked Services (DANS)Bloggers as experts
14 Citations2008Krisztian Balog, Maarten de Rijke +1 more
This work addresses the task of (blog) feed distillation: to find blogs that are principally devoted to a given topic, and adopts two language modeling-based approaches to expert finding, and determines their effectiveness as feeddistillation strategies.