Incorporating dwell time in session-based recommendations with recurrent Neural networks

Veronika Bogina; Tsvi Kuflik

Incorporating dwell time in session-based recommendations with recurrent Neural networks

Veronika Bogina, Tsvi Kuflik

Department of Information Systems

Research output: Contribution to journal › Conference article › peer-review

Abstract

Recurrent Neural Networks (RNN) is a frequently used technique for sequence data predictions. Recently, it gains popularity in the Recommender Systems domain, especially for session-based recommendations where naturally, each session is defined as a sequence of clicks, and timestamped data per click is available. In our research, in its early stages, we explore the value of incorporating dwell time into existing RNN framework for session-based recommendations by boosting items above the predefined dwell time threshold. We show improvement in recall@20 and MRR@20 by evaluating the proposed approach on e-commerce RecSys’15 challenge dataset.

Original language	English
Pages (from-to)	57-59
Number of pages	3
Journal	CEUR Workshop Proceedings
Volume	1922
State	Published - 2017
Event	1st Workshop on Temporal Reasoning in Recommender Systems, RecTemp 2017 - Como, Italy Duration: 27 Aug 2017 → 31 Aug 2017

Bibliographical note

Funding Information:
The work is partially supported by the Israeli Innovation Authority, the Ministry of Economy and Industry, MAGNET “Infomedia” project.

Publisher Copyright:
Copyright © 2017 for this paper by its authors.

Keywords

Deep learning
Dwell time
Recommender systems
Recurrent neural networks
Temporal aspects

ASJC Scopus subject areas

Computer Science (all)

Cite this

@article{2f86f0abc8a748d8bf6e0d5dba47bf70,

title = "Incorporating dwell time in session-based recommendations with recurrent Neural networks",

abstract = "Recurrent Neural Networks (RNN) is a frequently used technique for sequence data predictions. Recently, it gains popularity in the Recommender Systems domain, especially for session-based recommendations where naturally, each session is defined as a sequence of clicks, and timestamped data per click is available. In our research, in its early stages, we explore the value of incorporating dwell time into existing RNN framework for session-based recommendations by boosting items above the predefined dwell time threshold. We show improvement in recall@20 and MRR@20 by evaluating the proposed approach on e-commerce RecSys{\textquoteright}15 challenge dataset.",

keywords = "Deep learning, Dwell time, Recommender systems, Recurrent neural networks, Temporal aspects",

author = "Veronika Bogina and Tsvi Kuflik",

note = "Funding Information: The work is partially supported by the Israeli Innovation Authority, the Ministry of Economy and Industry, MAGNET “Infomedia” project. Publisher Copyright: Copyright {\textcopyright} 2017 for this paper by its authors.; 1st Workshop on Temporal Reasoning in Recommender Systems, RecTemp 2017 ; Conference date: 27-08-2017 Through 31-08-2017",

year = "2017",

language = "English",

volume = "1922",

pages = "57--59",

journal = "CEUR Workshop Proceedings",

issn = "1613-0073",

publisher = "CEUR-WS",

}

TY - JOUR

T1 - Incorporating dwell time in session-based recommendations with recurrent Neural networks

AU - Bogina, Veronika

AU - Kuflik, Tsvi

N1 - Funding Information: The work is partially supported by the Israeli Innovation Authority, the Ministry of Economy and Industry, MAGNET “Infomedia” project. Publisher Copyright: Copyright © 2017 for this paper by its authors.

PY - 2017

Y1 - 2017

N2 - Recurrent Neural Networks (RNN) is a frequently used technique for sequence data predictions. Recently, it gains popularity in the Recommender Systems domain, especially for session-based recommendations where naturally, each session is defined as a sequence of clicks, and timestamped data per click is available. In our research, in its early stages, we explore the value of incorporating dwell time into existing RNN framework for session-based recommendations by boosting items above the predefined dwell time threshold. We show improvement in recall@20 and MRR@20 by evaluating the proposed approach on e-commerce RecSys’15 challenge dataset.

AB - Recurrent Neural Networks (RNN) is a frequently used technique for sequence data predictions. Recently, it gains popularity in the Recommender Systems domain, especially for session-based recommendations where naturally, each session is defined as a sequence of clicks, and timestamped data per click is available. In our research, in its early stages, we explore the value of incorporating dwell time into existing RNN framework for session-based recommendations by boosting items above the predefined dwell time threshold. We show improvement in recall@20 and MRR@20 by evaluating the proposed approach on e-commerce RecSys’15 challenge dataset.

KW - Deep learning

KW - Dwell time

KW - Recommender systems

KW - Recurrent neural networks

KW - Temporal aspects

UR - http://www.scopus.com/inward/record.url?scp=85030320339&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85030320339

SN - 1613-0073

VL - 1922

SP - 57

EP - 59

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

T2 - 1st Workshop on Temporal Reasoning in Recommender Systems, RecTemp 2017

Y2 - 27 August 2017 through 31 August 2017

ER -

Incorporating dwell time in session-based recommendations with recurrent Neural networks

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this