Voting Advice Applications: Missing Value Estimation Using Matrix Factorization and Collaborative Filtering - Artificial Intelligence Applications and Innovations Access content directly
Conference Papers Year : 2013

Voting Advice Applications: Missing Value Estimation Using Matrix Factorization and Collaborative Filtering

Marilena Agathokleous
  • Function : Author
  • PersonId : 1000698
Nicolas Tsapatsoulis
  • Function : Author
  • PersonId : 991067

Abstract

A Voting Advice Application (VAA) is a web application that recommends to a voter the party or the candidate, who replied like him/her in an online questionnaire. Every question is responding to the political positions of each party. If the voter fails to answer some questions, it is likely the VAA to offer him/her the wrong candidate. Therefore, it is necessary to inspect the missing data (not answered questions) and try to estimate them. In this paper we formulate the VAA missing value problem and investigate several different approaches of collaborative filtering to tackle it. The evaluation of the proposed approaches was done by using the data obtained from the Cypriot presidential elections of February 2013 and the parliamentary elections in Greece in May, 2012. The corresponding datasets are made freely available to other researchers working in the areas of VAA and recommender systems through the Web.
Fichier principal
Vignette du fichier
978-3-642-41142-7_3_Chapter.pdf (225.51 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01459677 , version 1 (07-02-2017)

Licence

Attribution

Identifiers

Cite

Marilena Agathokleous, Nicolas Tsapatsoulis. Voting Advice Applications: Missing Value Estimation Using Matrix Factorization and Collaborative Filtering. 9th Artificial Intelligence Applications and Innovations (AIAI), Sep 2013, Paphos, Greece. pp.20-29, ⟨10.1007/978-3-642-41142-7_3⟩. ⟨hal-01459677⟩
59 View
664 Download

Altmetric

Share

Gmail Facebook X LinkedIn More