Detecting Document Genre for Personalization of Information Retrieval

Muresan, Gheorghe; Smith, Catherine L.; Cole, Michael; Liu, Lu; Belkin, Nicholas J.

doi:10.1109/hicss.2006.119

Cited by 7 publications

(4 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…La precisión mejora inversamente y la exhaustividad proporcionalmente al número de características unidas algebraicamente. Por otro lado, se ha descartado la hipótesis de que los géneros se puedan reconocer por el vocabulario empleado, pues la combinación de características lingüísticas generales resultó insuficiente para la clasificación de géneros (Muresan, et al, 2006). Otros estudios ponen de relieve que los clasificadores automáticos funcionan mejor cuando se tiene en cuenta la relación entre género y tema (Finn;Kushmerick, 2006;Myaeng, 2004).…”

Section: Reconocimiento Automático De Los Géneros Webunclassified

Géneros web: líneas de investigación

Montesi¹

2008

El Profesional de la Informacion

View full text Add to dashboard Cite

Section: Reconocimiento Automático De Los Géneros Webunclassified

Géneros web: líneas de investigación

Montesi¹

2008

El Profesional de la Informacion

View full text Add to dashboard Cite

“…Sharoff, 2007), information retrieval (e.g. Muresan et al, 2006), social network analysis (e.g. Paolillo et al, 2007), web mining (e.g.…”

Section: Introductionmentioning

confidence: 99%

Testing a Genre-Enabled Application: A Preliminary Assessment

Santín

Rosso

2008

Electronic Workshops in Computing

View full text Add to dashboard Cite

In this paper we would like to contribute to the discussion about genre-enabled applications, currently engaging many genre researchers, by presenting a preliminary assessment of a web add-on devised to augment the result list of general-purpose search engines with genre labels. For this assessment, we use a small collection of web pages manually annotated with genre labels by a large number of web users. This resource is made up of two sets of web pages created by two independent researchers for their own user-based genre studies. This comparison allows us to provide a preliminary view on the genre add-on performance and to highlight some open issues in genre research.

show abstract

“…No structural document features were included, due to the challenge of extracting these from documents in a wide range of file formats. Word stemming, which is common in subject-based classification, was not used as there is insufficient evidence to suggest that word stems are more expressive of genre than inflected terms (Muresan, Smith, Cole, Liu, & Belkin, 2006). SVM Light uses tuning parameters to set the rate of error tolerance and the relative weight of positive versus negative examples in the training set.…”

Section: Genre Classifiermentioning

confidence: 99%

Exploiting task-document relations in support of information retrieval in the workplace

Freund

2008

SIGIR Forum

View full text Add to dashboard Cite

Increasingly, workplace information seeking takes place in digital information environments and is reliant upon search systems. Existing systems are designed to retrieve information that is relevant to the query, but are not capable of identifying information that is well-suited to the context and situation of a search. This is a problem for professionals who often are searching for a small amount of useful information that can be applied to a problem or task, and have limited time to browse through large sets of results. This inability of search systems to discriminate between relevant and useful documents is one of the core problems in information retrieval.In this dissertation, I address this problem by studying the role that contextual factors play in determining how a group of professionals searches for and selects information. The central question concerns the nature of the relationships between these contextual factors, specifically between the genres in the document collection and the tasks of the searcher, with an aim to exploit such relationships to improve workplace information retrieval. Research was conducted through multiple studies in three phases, moving from an exploratory study of workplace information behaviour to a controlled experimental user study.Findings confirm that workplace context shapes search behaviour. This relationship is modeled as a set of key contextual factors and sets of context-dependent access constraints, preferred document characteristics, and search strategies. Among the contextual factors identified, work ii tasks and information tasks were found to be significantly associated with document genres. This task-genre relationship was modeled as a matrix of associations between domain-specific task and genre taxonomies and successfully implemented as a filtering component in a workplace search system. This is the first major study of the relationship between task and genre in information seeking and of its application to information retrieval systems.iii AcknowledgementsThis document has been a long time coming and many people have helped me along the path to its completion. First among these are my wonderful children, Alon, Jonathan, Gabriel and Ethan, who, despite the injustice of having to grow up with a mother who never seemed to be able to finish her homework, have consistently cheered me on and always provided hugs when needed. A great debt of thanks is also owed to my partner of many years, Amir, who encouraged me to take on this project, and who has supported our family through my many years as a student.

show abstract

Detecting Document Genre for Personalization of Information Retrieval

Cited by 7 publications

References 10 publications

Géneros web: líneas de investigación

Géneros web: líneas de investigación

Testing a Genre-Enabled Application: A Preliminary Assessment

Exploiting task-document relations in support of information retrieval in the workplace

Contact Info

Product

Resources

About