Unobtrusive dynamic modelling of TV programme preferences in a Finnish household

Elena Vildjiounaite (Corresponding Author), Vesa Kyllönen, Tero Hannula, Petteri Alahuhta

Research output: Contribution to journalArticleScientificpeer-review

18 Citations (Scopus)

Abstract

The majority of recommender systems require explicit user interaction (ranking of movies and TV programmes and/or their metadata, such as genres, actors etc), which requires user time and effort. Furthermore, such ranking is often done separately by each person, while merging these manually acquired individual preferences in multi-user environments remains largely an unsolved problem. This work presents a method for learning a joint model of a multi-user environment from implicit interactions: programme choices which family members make together and separately. The proposed method allows to adapt to the practices of each particular family and to protect family privacy, because the joint family model is learned for each family separately. Furthermore, since the accuracy of machine learning methods is family-dependent and none of the machine learning methods outperforms others for all families, a fairly lightweight classifier ensemble selection approach is applied for better adaptation to the specifics of each family. In tests on the real-life TV viewing histories of 20 families, acquired over 5 months, the classifier ensemble achieved an accuracy comparable with that of systems which require explicit user ratings: an average recall of 57% at an average precision of 30%, despite only a few programme metadata descriptors being available.
Original languageEnglish
Pages (from-to)143 - 157
Number of pages15
JournalMultimedia Systems
Volume15
Issue number3
DOIs
Publication statusPublished - 2009
MoE publication typeA1 Journal article-refereed

Fingerprint

Metadata
Learning systems
Classifiers
Recommender systems
Merging

Keywords

  • intelligent systems
  • TV recommender system
  • user modelling

Cite this

@article{4d2f8bfb91b941839ed570d39cbca3f2,
title = "Unobtrusive dynamic modelling of TV programme preferences in a Finnish household",
abstract = "The majority of recommender systems require explicit user interaction (ranking of movies and TV programmes and/or their metadata, such as genres, actors etc), which requires user time and effort. Furthermore, such ranking is often done separately by each person, while merging these manually acquired individual preferences in multi-user environments remains largely an unsolved problem. This work presents a method for learning a joint model of a multi-user environment from implicit interactions: programme choices which family members make together and separately. The proposed method allows to adapt to the practices of each particular family and to protect family privacy, because the joint family model is learned for each family separately. Furthermore, since the accuracy of machine learning methods is family-dependent and none of the machine learning methods outperforms others for all families, a fairly lightweight classifier ensemble selection approach is applied for better adaptation to the specifics of each family. In tests on the real-life TV viewing histories of 20 families, acquired over 5 months, the classifier ensemble achieved an accuracy comparable with that of systems which require explicit user ratings: an average recall of 57{\%} at an average precision of 30{\%}, despite only a few programme metadata descriptors being available.",
keywords = "intelligent systems, TV recommender system, user modelling",
author = "Elena Vildjiounaite and Vesa Kyll{\"o}nen and Tero Hannula and Petteri Alahuhta",
year = "2009",
doi = "10.1007/s00530-009-0157-1",
language = "English",
volume = "15",
pages = "143 -- 157",
journal = "Multimedia Systems",
issn = "0942-4962",
publisher = "Springer",
number = "3",

}

Unobtrusive dynamic modelling of TV programme preferences in a Finnish household. / Vildjiounaite, Elena (Corresponding Author); Kyllönen, Vesa; Hannula, Tero; Alahuhta, Petteri.

In: Multimedia Systems, Vol. 15, No. 3, 2009, p. 143 - 157.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Unobtrusive dynamic modelling of TV programme preferences in a Finnish household

AU - Vildjiounaite, Elena

AU - Kyllönen, Vesa

AU - Hannula, Tero

AU - Alahuhta, Petteri

PY - 2009

Y1 - 2009

N2 - The majority of recommender systems require explicit user interaction (ranking of movies and TV programmes and/or their metadata, such as genres, actors etc), which requires user time and effort. Furthermore, such ranking is often done separately by each person, while merging these manually acquired individual preferences in multi-user environments remains largely an unsolved problem. This work presents a method for learning a joint model of a multi-user environment from implicit interactions: programme choices which family members make together and separately. The proposed method allows to adapt to the practices of each particular family and to protect family privacy, because the joint family model is learned for each family separately. Furthermore, since the accuracy of machine learning methods is family-dependent and none of the machine learning methods outperforms others for all families, a fairly lightweight classifier ensemble selection approach is applied for better adaptation to the specifics of each family. In tests on the real-life TV viewing histories of 20 families, acquired over 5 months, the classifier ensemble achieved an accuracy comparable with that of systems which require explicit user ratings: an average recall of 57% at an average precision of 30%, despite only a few programme metadata descriptors being available.

AB - The majority of recommender systems require explicit user interaction (ranking of movies and TV programmes and/or their metadata, such as genres, actors etc), which requires user time and effort. Furthermore, such ranking is often done separately by each person, while merging these manually acquired individual preferences in multi-user environments remains largely an unsolved problem. This work presents a method for learning a joint model of a multi-user environment from implicit interactions: programme choices which family members make together and separately. The proposed method allows to adapt to the practices of each particular family and to protect family privacy, because the joint family model is learned for each family separately. Furthermore, since the accuracy of machine learning methods is family-dependent and none of the machine learning methods outperforms others for all families, a fairly lightweight classifier ensemble selection approach is applied for better adaptation to the specifics of each family. In tests on the real-life TV viewing histories of 20 families, acquired over 5 months, the classifier ensemble achieved an accuracy comparable with that of systems which require explicit user ratings: an average recall of 57% at an average precision of 30%, despite only a few programme metadata descriptors being available.

KW - intelligent systems

KW - TV recommender system

KW - user modelling

U2 - 10.1007/s00530-009-0157-1

DO - 10.1007/s00530-009-0157-1

M3 - Article

VL - 15

SP - 143

EP - 157

JO - Multimedia Systems

JF - Multimedia Systems

SN - 0942-4962

IS - 3

ER -