A User Experience Model for Privacy and Context Aware Over-the-Top (OTT) TV Recommendations

Valentino Servizi, Sokol Kosta, Allan Hammershoj, Henning Olesen

Publikation: Bidrag til tidsskriftKonferenceartikel i tidsskriftForskningpeer review

2 Downloads (Pure)

Resumé

Conventional recommender systems provide personalized recommendations by collecting and retaining user data, relying on a centralized architecture. Hence, user privacy is undermined by the volume of information required to support the personalized experience. In this work, we propose a User Experience model witch allows the privacy preservation of a user by a decentralized architecture, enabling the Service Provider to offer recommendations without the need of storing individual user data. We advance the current state of the art by: i) Proposing a model of User Experience suitable for Persona-based recommendations; ii) Presenting a User Experience collection model able to enhance the user privacy towards the service provider while keeping the quality of her preferences predictions; and iii) Assessing the existence of the Persona profiles, which are needed for generating and addressing the recommendations. We perform several experiments using a real-world complete dataset from a medium-sized service provider, composed of more than 14,000 unique users and 33,000 content titles collected over a period of two years. We show that our architecture, in com- bination with our User Experience model, achieves the same results or better, in terms of rating prediction accuracy, as the state of the art systems, without sacrificing user’s privacy.
OriginalsprogEngelsk
TidsskriftCEUR Workshop Proceedings
Vol/bind2482
Antal sider8
ISSN1613-0073
StatusUdgivet - 2019
Begivenhed27th ACM International Conference on Information and Knowledge Management (CIKM 2018) - Torino, Italien
Varighed: 22 okt. 201822 okt. 2018

Konference

Konference27th ACM International Conference on Information and Knowledge Management (CIKM 2018)
LandItalien
ByTorino
Periode22/10/201822/10/2018

Fingerprint

Recommender systems
Experiments

Citer dette

@inproceedings{ae0350f723674acc8ac029451e773527,
title = "A User Experience Model for Privacy and Context Aware Over-the-Top (OTT) TV Recommendations",
abstract = "Conventional recommender systems provide personalized recommendations by collecting and retaining user data, relying on a centralized architecture. Hence, user privacy is undermined by the volume of information required to support the personalized experience. In this work, we propose a User Experience model witch allows the privacy preservation of a user by a decentralized architecture, enabling the Service Provider to offer recommendations without the need of storing individual user data. We advance the current state of the art by: i) Proposing a model of User Experience suitable for Persona-based recommendations; ii) Presenting a User Experience collection model able to enhance the user privacy towards the service provider while keeping the quality of her preferences predictions; and iii) Assessing the existence of the Persona profiles, which are needed for generating and addressing the recommendations. We perform several experiments using a real-world complete dataset from a medium-sized service provider, composed of more than 14,000 unique users and 33,000 content titles collected over a period of two years. We show that our architecture, in com- bination with our User Experience model, achieves the same results or better, in terms of rating prediction accuracy, as the state of the art systems, without sacrificing user’s privacy.",
author = "Valentino Servizi and Sokol Kosta and Allan Hammershoj and Henning Olesen",
year = "2019",
language = "English",
volume = "2482",
journal = "CEUR Workshop Proceedings",
issn = "1613-0073",
publisher = "CEUR Workshop Proceedings",

}

A User Experience Model for Privacy and Context Aware Over-the-Top (OTT) TV Recommendations. / Servizi, Valentino; Kosta, Sokol; Hammershoj, Allan; Olesen, Henning.

I: CEUR Workshop Proceedings, Bind 2482, 2019.

Publikation: Bidrag til tidsskriftKonferenceartikel i tidsskriftForskningpeer review

TY - GEN

T1 - A User Experience Model for Privacy and Context Aware Over-the-Top (OTT) TV Recommendations

AU - Servizi, Valentino

AU - Kosta, Sokol

AU - Hammershoj, Allan

AU - Olesen, Henning

PY - 2019

Y1 - 2019

N2 - Conventional recommender systems provide personalized recommendations by collecting and retaining user data, relying on a centralized architecture. Hence, user privacy is undermined by the volume of information required to support the personalized experience. In this work, we propose a User Experience model witch allows the privacy preservation of a user by a decentralized architecture, enabling the Service Provider to offer recommendations without the need of storing individual user data. We advance the current state of the art by: i) Proposing a model of User Experience suitable for Persona-based recommendations; ii) Presenting a User Experience collection model able to enhance the user privacy towards the service provider while keeping the quality of her preferences predictions; and iii) Assessing the existence of the Persona profiles, which are needed for generating and addressing the recommendations. We perform several experiments using a real-world complete dataset from a medium-sized service provider, composed of more than 14,000 unique users and 33,000 content titles collected over a period of two years. We show that our architecture, in com- bination with our User Experience model, achieves the same results or better, in terms of rating prediction accuracy, as the state of the art systems, without sacrificing user’s privacy.

AB - Conventional recommender systems provide personalized recommendations by collecting and retaining user data, relying on a centralized architecture. Hence, user privacy is undermined by the volume of information required to support the personalized experience. In this work, we propose a User Experience model witch allows the privacy preservation of a user by a decentralized architecture, enabling the Service Provider to offer recommendations without the need of storing individual user data. We advance the current state of the art by: i) Proposing a model of User Experience suitable for Persona-based recommendations; ii) Presenting a User Experience collection model able to enhance the user privacy towards the service provider while keeping the quality of her preferences predictions; and iii) Assessing the existence of the Persona profiles, which are needed for generating and addressing the recommendations. We perform several experiments using a real-world complete dataset from a medium-sized service provider, composed of more than 14,000 unique users and 33,000 content titles collected over a period of two years. We show that our architecture, in com- bination with our User Experience model, achieves the same results or better, in terms of rating prediction accuracy, as the state of the art systems, without sacrificing user’s privacy.

UR - http://ceur-ws.org/Vol-2482/paper45.pdf

M3 - Conference article in Journal

VL - 2482

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

SN - 1613-0073

ER -