Suchergebnisse

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 election 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 similarity-based approach performs well in terms not only of recommendation accuracy but also efficiency (we obtain smaller profiles and consequently faster recommendations). ; This work has been funded by the Spanish Ministerio de Economía y Competitividad under projects TIN2013-42741-P and TIN2016-77902-C3-2-P, and the European Regional Development Fund ( ERDF-FEDER ).

One step towards breaking down barriers between citizens and politicians is to help people identify those politicians who share their concerns. This paper is set in the field of expert finding and is based on the automatic construction of politicians' profiles from their speeches on parliamentary committees. These committee-based profiles are treated as documents and are indexed by an information retrieval system. Given a query representing a citizen's concern, a profile ranking is then obtained. In the final step, the different results for each candidate are combined in order to obtain the final politician ranking. We explore the use of classic combination strategies for this purpose and present a new approach that improves state-of-the-art performance and which is more stable under different conditions. We also introduce a two-stage model where the identification of a broader concept (such as the committee) is used to improve the final politician ranking. ; This work has been funded by the Spanish Ministerio de Economı́a y Competitividad under projects TIN2013-42741-P and TIN2016-77902-C3-2-P, and the European Regional Development Fund (ERDF-FEDER).

In the context of e-government and more specifically that of parliament, this paper tackles the problem of finding Members of Parliament (MPs) according to their profiles which have been built from their speeches in plenary or committee sessions. The paper presents a common solution for two problems: firstly, a member of the public who is concerned about a certain issue might want to know who the best MP is for dealing with their problem (recommending task); and secondly, each new piece of textual information that reaches the house must be correctly allocated to the appropriate MP according to its content (filtering task). This paper explores both these ways of searching for relevant people conceptually by encapsulating them into a single problem: that of searching for the relevant MP's profile given an information need. Our research work proposes various profile construction methods (by selecting and weighting appropriate terms) and compares these using different retrieval models to evaluate their quality and suitability for different types of information needs in order to simulate real and common situations. ; This research has been supported by the Spanish 'Ministerio de Economı́a y Competitividad' under project TIN2013-42741-P and the European Regional Development Fund.

Our goal is to learn about the political interests and preferences of Members of Parliament (MPs) by mining their parliamentary activity in order to develop a recommendation/filtering system to determine how relevant documents should be distributed among MPs. We propose the use of positive unlabeled learning to tackle this problem since we only have information about relevant documents (the interventions of each MP in debates) but not about irrelevant documents and so it is not possible to use standard binary classifiers which have been trained with positive and negative examples. Additionally, we have also developed a new positive unlabeled learning algorithm that compares favorably with: (a) a baseline approach which assumes that every intervention by any other MP is irrelevant; (b) another well-known positive unlabeled learning method; and (c) an approach based on information retrieval methods that matches documents and legislators' representations. The experiments have been conducted with data from the regional Spanish Andalusian Parliament. ; This work has been funded by the Spanish "Ministerio de Economía y Competitividad" under projects TIN2013-42741-P and TIN2016-77902-C3-2-P, and the European Regional Development Fund (ERDF-FEDER).