Posts tagged Relevance modeling

TREC

The University of Amsterdam at Trec 2010: Session, Entity, and Relevance Feedback

0
1 year ago
by in ,

We describe the par­tic­i­pa­tion of the Uni­ver­sity of Amsterdam’s ILPS group in the ses­sion, entity, and rel­e­vance feed­back track at TREC 2010. In the Ses­sion Track we explore the use of blind rel­e­vance feed­back to bias a follow-up query towards or against the top­ics cov­ered in doc­u­ments returned More >

thesis cover image of a smart computer

Combining Concepts and Language Models for Information Access

0
1 year ago
by in ,

Since the mid­dle of last cen­tury, infor­ma­tion retrieval has gained an increas­ing inter­est. Since its incep­tion, much research has been devoted to find­ing opti­mal ways of rep­re­sent­ing both doc­u­ments and queries, as well as improv­ing ways of match­ing one with the other. In cases where doc­u­ment anno­ta­tions or explicit seman­tics More >

TREC

Topical Diversity and Relevance Feedback

0
2 years ago
by in ,

We describe the par­tic­i­pa­tion of the Uni­ver­sity of Amsterdam’s Intel­li­gent Sys­tems Lab in the rel­e­vance feed­back track at TREC 2009. Our main con­clu­sion for the rel­e­vance feed­back track is that a top­i­cal diver­sity approach pro­vides good feed­back doc­u­ments. Fur­ther, we find that our rel­e­vance feed­back algo­rithm seems to help More >

formula

A query model based on normalized log-likelihood

0
2 years ago
by in ,

A query is usu­ally a brief, some­times impre­cise expres­sion of an under­ly­ing infor­ma­tion need . Exam­in­ing how queries can be trans­formed to equiv­a­lent, poten­tially bet­ter queries is a theme of recur­ring inter­est to the infor­ma­tion retrieval com­mu­nity. Such trans­for­ma­tions include expan­sion of short queries to long queries, para­phras­ing queries using More >

Stack of books

Concept models for domain-specific search

0
3 years ago
by in ,

We describe our par­tic­i­pa­tion in the 2008 CLEF Domain-specific track. We eval­u­ate blind rel­e­vance feed­back mod­els and con­cept mod­els on the CLEF domain-specific test col­lec­tion. Apply­ing rel­e­vance mod­el­ing tech­niques is found to have a pos­i­tive effect on the 2008 topic set, in terms of mean aver­age pre­ci­sion and More >

CLEF domain-specific sample graphic

The University of Amsterdam at the CLEF 2008 Domain Specific Track — Parsimonious Relevance and Concept Models

0
3 years ago
by in ,

We describe our par­tic­i­pa­tion in the CLEF 2008 Domain Spe­cific track. The research ques­tions we address are three­fold: (i) what are the effects of esti­mat­ing and apply­ing rel­e­vance mod­els to the domain spe­cific col­lec­tion used at CLEF 2008, (ii) what are the results of par­si­mo­niz­ing these rel­e­vance mod­els, More >

mathematics

Parsimonious Relevance Models

0
3 years ago
by in ,

Rel­e­vance feed­back is often applied to bet­ter cap­ture a user’s infor­ma­tion need. Auto­mat­i­cally refor­mu­lat­ing queries (or blind rel­e­vance feed­back) entails look­ing at the terms in some set of (pseudo-)relevant doc­u­ments and select­ing the most infor­ma­tive ones with respect to the set or the col­lec­tion. These terms may then be reweighed More >

dna_1

Thesaurus-Based Feedback to Support Mixed Search and Browsing Environments

0
4 years ago
by in ,

We pro­pose and eval­u­ate a query expan­sion mech­a­nism that sup­ports search­ing and brows­ing in col­lec­tions of anno­tated doc­u­ments. Based on gen­er­a­tive lan­guage mod­els, our feed­back mech­a­nism uses document-level anno­ta­tions to bias the gen­er­a­tion of expan­sion terms and to gen­er­ate brows­ing sug­ges­tions in the form of con­cepts selected from a con­trolled More >

Go to Top