Unobtrusive Dynamic Modelling of TV Program Preferences in a Household

Elena Vildjiounaite; Vesa Kyllönen; Tero Hannula; Petteri Alahuhta

doi:10.1007/978-3-540-69478-6_9

Unobtrusive Dynamic Modelling of TV Program Preferences in a Household

Elena Vildjiounaite, Vesa Kyllönen, Tero Hannula, Petteri Alahuhta

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

13 Citations (Scopus)

Abstract

Majority of recommender systems require explicit user interaction (ranking of movies and TV programs and/or their metadata, such as genres, actors etc), which requires user time and effort. Furthermore, often such ranking is done separately by each person, while merging these manually acquired preferences in multi-user environments remains largely unsolved problem. This work presents a method to learn a model of multi-user environment in intelligent home from implicit interactions: the choices which family members make together and separately. In tests on TV viewing histories of twenty families, acquired during two months, the method has achieved prediction accuracy comparable with the accuracy of systems which require explicit user ratings: a set of TV programs, actually viewed during each test session (average set size was 2.2 programs per viewing session), was recommended among five top choices in 60% of cases on average, despite training on small data sets.

Original language	English
Title of host publication	Changing Television Environments
Subtitle of host publication	6th European Conference, EUROITV 2008, Proceedings
Editors	Manfred Tscheligi, Marianna Obrist, Artur Lugmayr
Publisher	Springer
Pages	82-91
ISBN (Electronic)	978-3-540-69478-6
ISBN (Print)	978-3-540-69477-9
DOIs	https://doi.org/10.1007/978-3-540-69478-6_9
Publication status	Published - 2008
MoE publication type	A4 Article in a conference publication
Event	6th European Conference, EUROITV 2008 - Salzburg, Austria Duration: 3 Jul 2008 → 4 Jul 2008

Publication series

Series	Lecture Notes in Computer Science
Volume	5066
ISSN	0302-9743

Conference

Conference	6th European Conference, EUROITV 2008
Abbreviated title	EUROITV 2008
Country	Austria
City	Salzburg
Period	3/07/08 → 4/07/08

Keywords

smart home
TV recommender system
user modelling
intelligent user service
user needs

Access to Document

10.1007/978-3-540-69478-6_9

Fingerprint Dive into the research topics of 'Unobtrusive Dynamic Modelling of TV Program Preferences in a Household'. Together they form a unique fingerprint.

Cite this

Vildjiounaite, E., Kyllönen, V., Hannula, T., & Alahuhta, P. (2008). Unobtrusive Dynamic Modelling of TV Program Preferences in a Household. In M. Tscheligi, M. Obrist, & A. Lugmayr (Eds.), Changing Television Environments: 6th European Conference, EUROITV 2008, Proceedings (pp. 82-91). Springer. Lecture Notes in Computer Science, Vol.. 5066 https://doi.org/10.1007/978-3-540-69478-6_9

Vildjiounaite, Elena ; Kyllönen, Vesa ; Hannula, Tero ; Alahuhta, Petteri. / Unobtrusive Dynamic Modelling of TV Program Preferences in a Household. Changing Television Environments: 6th European Conference, EUROITV 2008, Proceedings. editor / Manfred Tscheligi ; Marianna Obrist ; Artur Lugmayr. Springer, 2008. pp. 82-91 (Lecture Notes in Computer Science, Vol. 5066).

@inproceedings{057b0a08143b40b09838b5fcc207fd7c,

title = "Unobtrusive Dynamic Modelling of TV Program Preferences in a Household",

abstract = "Majority of recommender systems require explicit user interaction (ranking of movies and TV programs and/or their metadata, such as genres, actors etc), which requires user time and effort. Furthermore, often such ranking is done separately by each person, while merging these manually acquired preferences in multi-user environments remains largely unsolved problem. This work presents a method to learn a model of multi-user environment in intelligent home from implicit interactions: the choices which family members make together and separately. In tests on TV viewing histories of twenty families, acquired during two months, the method has achieved prediction accuracy comparable with the accuracy of systems which require explicit user ratings: a set of TV programs, actually viewed during each test session (average set size was 2.2 programs per viewing session), was recommended among five top choices in 60% of cases on average, despite training on small data sets.",

keywords = "smart home, TV recommender system, user modelling, intelligent user service, user needs",

author = "Elena Vildjiounaite and Vesa Kyll{\"o}nen and Tero Hannula and Petteri Alahuhta",

year = "2008",

doi = "10.1007/978-3-540-69478-6_9",

language = "English",

isbn = "978-3-540-69477-9",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "82--91",

editor = "Manfred Tscheligi and Obrist, {Marianna } and Artur Lugmayr",

booktitle = "Changing Television Environments",

address = "Germany",

note = "6th European Conference, EUROITV 2008, EUROITV 2008 ; Conference date: 03-07-2008 Through 04-07-2008",

}

Vildjiounaite, E , Kyllönen, V, Hannula, T & Alahuhta, P 2008, Unobtrusive Dynamic Modelling of TV Program Preferences in a Household. in M Tscheligi, M Obrist & A Lugmayr (eds), Changing Television Environments: 6th European Conference, EUROITV 2008, Proceedings. Springer, Lecture Notes in Computer Science, vol. 5066, pp. 82-91, 6th European Conference, EUROITV 2008, Salzburg, Austria, 3/07/08. https://doi.org/10.1007/978-3-540-69478-6_9

Unobtrusive Dynamic Modelling of TV Program Preferences in a Household. / Vildjiounaite, Elena ; Kyllönen, Vesa; Hannula, Tero; Alahuhta, Petteri.

Changing Television Environments: 6th European Conference, EUROITV 2008, Proceedings. ed. / Manfred Tscheligi; Marianna Obrist; Artur Lugmayr. Springer, 2008. p. 82-91 (Lecture Notes in Computer Science, Vol. 5066).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - Unobtrusive Dynamic Modelling of TV Program Preferences in a Household

AU - Vildjiounaite, Elena

AU - Kyllönen, Vesa

AU - Hannula, Tero

AU - Alahuhta, Petteri

PY - 2008

Y1 - 2008

N2 - Majority of recommender systems require explicit user interaction (ranking of movies and TV programs and/or their metadata, such as genres, actors etc), which requires user time and effort. Furthermore, often such ranking is done separately by each person, while merging these manually acquired preferences in multi-user environments remains largely unsolved problem. This work presents a method to learn a model of multi-user environment in intelligent home from implicit interactions: the choices which family members make together and separately. In tests on TV viewing histories of twenty families, acquired during two months, the method has achieved prediction accuracy comparable with the accuracy of systems which require explicit user ratings: a set of TV programs, actually viewed during each test session (average set size was 2.2 programs per viewing session), was recommended among five top choices in 60% of cases on average, despite training on small data sets.

AB - Majority of recommender systems require explicit user interaction (ranking of movies and TV programs and/or their metadata, such as genres, actors etc), which requires user time and effort. Furthermore, often such ranking is done separately by each person, while merging these manually acquired preferences in multi-user environments remains largely unsolved problem. This work presents a method to learn a model of multi-user environment in intelligent home from implicit interactions: the choices which family members make together and separately. In tests on TV viewing histories of twenty families, acquired during two months, the method has achieved prediction accuracy comparable with the accuracy of systems which require explicit user ratings: a set of TV programs, actually viewed during each test session (average set size was 2.2 programs per viewing session), was recommended among five top choices in 60% of cases on average, despite training on small data sets.

KW - smart home

KW - TV recommender system

KW - user modelling

KW - intelligent user service

KW - user needs

U2 - 10.1007/978-3-540-69478-6_9

DO - 10.1007/978-3-540-69478-6_9

M3 - Conference article in proceedings

SN - 978-3-540-69477-9

T3 - Lecture Notes in Computer Science

SP - 82

EP - 91

BT - Changing Television Environments

A2 - Tscheligi, Manfred

A2 - Obrist, Marianna

A2 - Lugmayr, Artur

PB - Springer

T2 - 6th European Conference, EUROITV 2008

Y2 - 3 July 2008 through 4 July 2008

ER -

Vildjiounaite E , Kyllönen V, Hannula T, Alahuhta P. Unobtrusive Dynamic Modelling of TV Program Preferences in a Household. In Tscheligi M, Obrist M, Lugmayr A, editors, Changing Television Environments: 6th European Conference, EUROITV 2008, Proceedings. Springer. 2008. p. 82-91. (Lecture Notes in Computer Science, Vol. 5066). https://doi.org/10.1007/978-3-540-69478-6_9