Countering contextual bias in TV watching behavior: introducing social trend as external contextual factor in TV recommenders

Jing Yuan, Felix Lorenz, Andreas Lommatzsch, Mu Mu, Nicholas Race, Frank Hopfgartner, Sahin Albayrak

Research output: Contribution to Book/Report typesConference contributionResearchpeer-review

Abstract

Context-awareness has become a critical factor to improve the prediction of user interest in modern online TV recommendation systems. In addition to individual user preferences, existing context-aware approaches like tensor factorization incorporate system-level contextual bias to increase predicting accuracy. We analyzed a user interaction dataset from a WebTV platform and found that such contextual bias creates a skewed selection of recommended programs which ultimately limits users in a filter bubble. To address this issue, we introduce the Twitter social stream as an external contextual factor to extend the choice with items related to social media events. We apply two trend indicators, Trend Momentum and SigniScore, to the Twitter histories of relevant programs. The evaluation reveals that Trend Momentum outperforms SigniScore and signalizes 96% of all peaks ahead of time regarding the selected candidate program titles.
Original languageEnglish
Title of host publicationACM International Conference on Interactive Experiences for Television and Online Video (ACM TVX)
Place of PublicationThe Netherlands
PublisherACM Press
Pages21-30
Number of pages10
ISBN (Print)978-1-4503-4529-3
DOIs
Publication statusPublished - 16 Jun 2017

Fingerprint

Momentum
Recommender systems
Factorization
Tensors

Keywords

  • Privacy reserving recommender
  • context-aware applications
  • event detection
  • userexperience
  • video on demand

Cite this

Yuan, J., Lorenz, F., Lommatzsch, A., Mu, M., Race, N., Hopfgartner, F., & Albayrak, S. (2017). Countering contextual bias in TV watching behavior: introducing social trend as external contextual factor in TV recommenders. In ACM International Conference on Interactive Experiences for Television and Online Video (ACM TVX) (pp. 21-30 ). The Netherlands: ACM Press. https://doi.org/10.1145/3077548.3077552
Yuan, Jing ; Lorenz, Felix ; Lommatzsch, Andreas ; Mu, Mu ; Race, Nicholas ; Hopfgartner, Frank ; Albayrak, Sahin. / Countering contextual bias in TV watching behavior: introducing social trend as external contextual factor in TV recommenders. ACM International Conference on Interactive Experiences for Television and Online Video (ACM TVX). The Netherlands : ACM Press, 2017. pp. 21-30
@inproceedings{9bc2cc998cd7475283a2d4e8cffaf3d9,
title = "Countering contextual bias in TV watching behavior: introducing social trend as external contextual factor in TV recommenders",
abstract = "Context-awareness has become a critical factor to improve the prediction of user interest in modern online TV recommendation systems. In addition to individual user preferences, existing context-aware approaches like tensor factorization incorporate system-level contextual bias to increase predicting accuracy. We analyzed a user interaction dataset from a WebTV platform and found that such contextual bias creates a skewed selection of recommended programs which ultimately limits users in a filter bubble. To address this issue, we introduce the Twitter social stream as an external contextual factor to extend the choice with items related to social media events. We apply two trend indicators, Trend Momentum and SigniScore, to the Twitter histories of relevant programs. The evaluation reveals that Trend Momentum outperforms SigniScore and signalizes 96{\%} of all peaks ahead of time regarding the selected candidate program titles.",
keywords = "Privacy reserving recommender, context-aware applications, event detection, userexperience, video on demand",
author = "Jing Yuan and Felix Lorenz and Andreas Lommatzsch and Mu Mu and Nicholas Race and Frank Hopfgartner and Sahin Albayrak",
year = "2017",
month = "6",
day = "16",
doi = "10.1145/3077548.3077552",
language = "English",
isbn = "978-1-4503-4529-3",
pages = "21--30",
booktitle = "ACM International Conference on Interactive Experiences for Television and Online Video (ACM TVX)",
publisher = "ACM Press",

}

Yuan, J, Lorenz, F, Lommatzsch, A, Mu, M, Race, N, Hopfgartner, F & Albayrak, S 2017, Countering contextual bias in TV watching behavior: introducing social trend as external contextual factor in TV recommenders. in ACM International Conference on Interactive Experiences for Television and Online Video (ACM TVX). ACM Press, The Netherlands, pp. 21-30 . https://doi.org/10.1145/3077548.3077552

Countering contextual bias in TV watching behavior: introducing social trend as external contextual factor in TV recommenders. / Yuan, Jing; Lorenz, Felix; Lommatzsch, Andreas; Mu, Mu; Race, Nicholas; Hopfgartner, Frank; Albayrak, Sahin.

ACM International Conference on Interactive Experiences for Television and Online Video (ACM TVX). The Netherlands : ACM Press, 2017. p. 21-30 .

Research output: Contribution to Book/Report typesConference contributionResearchpeer-review

TY - GEN

T1 - Countering contextual bias in TV watching behavior: introducing social trend as external contextual factor in TV recommenders

AU - Yuan, Jing

AU - Lorenz, Felix

AU - Lommatzsch, Andreas

AU - Mu, Mu

AU - Race, Nicholas

AU - Hopfgartner, Frank

AU - Albayrak, Sahin

PY - 2017/6/16

Y1 - 2017/6/16

N2 - Context-awareness has become a critical factor to improve the prediction of user interest in modern online TV recommendation systems. In addition to individual user preferences, existing context-aware approaches like tensor factorization incorporate system-level contextual bias to increase predicting accuracy. We analyzed a user interaction dataset from a WebTV platform and found that such contextual bias creates a skewed selection of recommended programs which ultimately limits users in a filter bubble. To address this issue, we introduce the Twitter social stream as an external contextual factor to extend the choice with items related to social media events. We apply two trend indicators, Trend Momentum and SigniScore, to the Twitter histories of relevant programs. The evaluation reveals that Trend Momentum outperforms SigniScore and signalizes 96% of all peaks ahead of time regarding the selected candidate program titles.

AB - Context-awareness has become a critical factor to improve the prediction of user interest in modern online TV recommendation systems. In addition to individual user preferences, existing context-aware approaches like tensor factorization incorporate system-level contextual bias to increase predicting accuracy. We analyzed a user interaction dataset from a WebTV platform and found that such contextual bias creates a skewed selection of recommended programs which ultimately limits users in a filter bubble. To address this issue, we introduce the Twitter social stream as an external contextual factor to extend the choice with items related to social media events. We apply two trend indicators, Trend Momentum and SigniScore, to the Twitter histories of relevant programs. The evaluation reveals that Trend Momentum outperforms SigniScore and signalizes 96% of all peaks ahead of time regarding the selected candidate program titles.

KW - Privacy reserving recommender

KW - context-aware applications

KW - event detection

KW - userexperience

KW - video on demand

U2 - 10.1145/3077548.3077552

DO - 10.1145/3077548.3077552

M3 - Conference contribution

SN - 978-1-4503-4529-3

SP - 21

EP - 30

BT - ACM International Conference on Interactive Experiences for Television and Online Video (ACM TVX)

PB - ACM Press

CY - The Netherlands

ER -

Yuan J, Lorenz F, Lommatzsch A, Mu M, Race N, Hopfgartner F et al. Countering contextual bias in TV watching behavior: introducing social trend as external contextual factor in TV recommenders. In ACM International Conference on Interactive Experiences for Television and Online Video (ACM TVX). The Netherlands: ACM Press. 2017. p. 21-30 https://doi.org/10.1145/3077548.3077552