Use of textual and conceptual profiles for personalized retrieval of political documents

Vicente-López, Eduardo; Campos, Luis M. de; Fernández‐Luna, Juan M.; Huete, Juan F.

doi:10.1016/j.knosys.2016.09.005

Cited by 10 publications

(6 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When we consider personalization, a third component comes into play, the user profile. There are different ways to collect and represent user information [28]. In this article, we shall work with user profiles which are represented as a set of weighted keywords.…”

Section: Including the Profile In Predictorsmentioning

confidence: 99%

“…Each user submitted one or several of the previous 23 queries to the IRS assuming, among a fixed set of generic profiles, the profile(s) that best fits them. These generic user profiles were automatically learned from the content of the documents in each committee session and were represented as sets of weighted terms [28]. There are eight different generic user profiles relating to administration, agriculture, culture, economy, education, employment, environment and health.…”

Section: Experimental Frameworkmentioning

confidence: 99%

See 1 more Smart Citation

Predicting IR personalization performance using pre-retrieval query predictors

et al. 2018

Self Cite

View full text Add to dashboard Cite

Although personalization generally improves query performance, it may also occasionally harm how queries perform. If we are able to predict and therefore disable personalization for such situations, overall performance will be higher and users will be more satisfied with personalized systems. We use various state-ofthe-art, pre-retrieval query performance predictors and propose several others including user profile information for this purpose. We study the correlations between these predictors and the difference between personalized and original queries. We also use classification and regression techniques to improve the results and finally achieve slightly more than one third of maximum ideal performance. We consider this to be a good starting point within this research line, which will undoubtedly result in further work and improvements.

show abstract

Section: Including the Profile In Predictorsmentioning

confidence: 99%

Section: Experimental Frameworkmentioning

confidence: 99%

Predicting IR personalization performance using pre-retrieval query predictors

et al. 2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…• Difference (Diff) was introduced in [51] in the context of personalized search and was one of the weighting schemes used in [16] for building MP profiles, and has certain similarities to the relative document frequency proposed in [38]. The Diff measure of a term t for a document D j is the normalized frequency of t in D j minus the normalized frequency of t outside D j (i.e.…”

Section: Weighting Measuresmentioning

confidence: 99%

On the selection of the correct number of terms for profile construction: Theoretical and empirical analysis

Campos

Fernández‐Luna

Huete

2018

Information Sciences

Self Cite

View full text Add to dashboard Cite

In this paper, we examine the problem of building a user profile from a set of documents. This profile will consist of a subset of the most representative terms in the documents that best represent user preferences or interests. Inspired by the discrete concentration theory we have conducted an axiomatic study of seven properties that a selection function should fulfill: the minimum and maximum uncertainty principle, invariant to adding zeros, invariant to scale transformations, principle of nominal increase, transfer principle and the richest get richer inequality. We also present a novel selection function based on the use of similarity metrics, and more specifically the cosine measure which is commonly used in information retrieval, and demonstrate that this verifies six of the properties in addition to a weaker variant of the transfer principle, thereby representing a good selection approach.The theoretical study was complemented with an empirical study to compare the performance of different selection criteria (weight-and unweight-based) using real data in a parliamentary setting. In this study, we analyze the performance of the different functions focusing on the two main factors affecting the selection process: profile size (number of terms) and weight distribution. These profiles are then used in a document filtering task to show that our similaritybased approach performs well in terms not only of recommendation accuracy but also efficiency (we obtain smaller profiles and consequently faster recommendations).

show abstract

“…Vol. 32 According to [23,24] in the construction of a user profile, three fundamental phases related to collecting, constructing, and using the data acquired by analyzing the user's behavior in a CRS are identified. IRSs register the user's actions to collect as much information as possible to identify behavioral patterns and obtain their preferences.…”

Section: Introductionmentioning

confidence: 99%

Modelo para la recuperación de información con expansión de consulta y perfil de preferencia de los usuarios

Viltres-Sala,

Estrada-Sentí,

Febles-Rodríguez

et al. 2023

Rev. Fac. Ing.

View full text Add to dashboard Cite

Comprender la intención de búsqueda del usuario permite identificar y extraer los resultados de búsqueda más relevantes y personalizados de la información disponible según sus necesidades. En el presente artículo se plantea un algoritmo para la recuperación de información relevante que combina las preferencias del perfil del usuario y la expansión de consulta para obtener resultados de búsqueda relevantes y personalizados. El proceso de recuperación de información se valida mediante las métricas de Precision, Recall y Mean Average Precision (MAP) aplicadas a un conjunto de datos que contiene los documentos estandarizados y los perfiles de preferencias. Los resultados permitieron demostrar que el algoritmo mejora el proceso de recuperación de información al arrojar documentos con mejor calidad y relevancia según las necesidades de los usuarios.

show abstract

Use of textual and conceptual profiles for personalized retrieval of political documents

Cited by 10 publications

References 33 publications

Predicting IR personalization performance using pre-retrieval query predictors

Predicting IR personalization performance using pre-retrieval query predictors

On the selection of the correct number of terms for profile construction: Theoretical and empirical analysis

Modelo para la recuperación de información con expansión de consulta y perfil de preferencia de los usuarios

Contact Info

Product

Resources

About