Personalized Reason Generation for Explainable Song Recommendation

Guoshuai Zhao, Hao Fu, Ruihua Song, Tetsuya Sakai, Zhongxia Chen, Xing Xie, Xueming Qian

Research output: Contribution to journalArticle

Abstract

Personalized recommendation has received a lot of attention as a highly practical research topic. However, existing recommender systems provide the recommendations with a generic statement such as "Customers who bought this item also bought . . . ". Explainable recommendation, which makes a user aware of why such items are recommended, is in demand. The goal of our research is to make the users feel as if they are receiving recommendations from their friends. To this end, we formulate a new challenging problem called personalized reason generation for explainable recommendation for songs in conversation applications and propose a solution that generates a natural language explanation of the reason for recommending a song to that particular user. For example, if the user is a student, our method can generate an output such as "Campus radio plays this song at noon every day, and I think it sounds wonderful," which the student may find easy to relate to. In the offline experiments, through manual assessments, the gain of our method is statistically significant on the relevance to songs and personalization to users comparing with baselines. Large-scale online experiments show that our method outperforms manually selected reasons by 8.2% in terms of click-Through rate. Evaluation results indicate that our generated reasons are relevant to songs and personalized to users, and they attract users to click the recommendations.

Original languageEnglish
Article number41
JournalACM Transactions on Intelligent Systems and Technology
Volume10
Issue number4
DOIs
Publication statusPublished - 2019 Jul 1

Fingerprint

Recommendations
Students
Recommender systems
Experiments
Acoustic waves
Personalized Recommendation
Recommender Systems
Personalization
Natural Language
Experiment
Baseline
Customers
Output
Evaluation

Keywords

  • Conversational recommendation
  • explainable recommendation
  • natural language generation
  • personalization
  • recommender system

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Artificial Intelligence

Cite this

Personalized Reason Generation for Explainable Song Recommendation. / Zhao, Guoshuai; Fu, Hao; Song, Ruihua; Sakai, Tetsuya; Chen, Zhongxia; Xie, Xing; Qian, Xueming.

In: ACM Transactions on Intelligent Systems and Technology, Vol. 10, No. 4, 41, 01.07.2019.

Research output: Contribution to journalArticle

Zhao, Guoshuai ; Fu, Hao ; Song, Ruihua ; Sakai, Tetsuya ; Chen, Zhongxia ; Xie, Xing ; Qian, Xueming. / Personalized Reason Generation for Explainable Song Recommendation. In: ACM Transactions on Intelligent Systems and Technology. 2019 ; Vol. 10, No. 4.
@article{40eb059d724c47d59a6f20fbb8e93c9d,
title = "Personalized Reason Generation for Explainable Song Recommendation",
abstract = "Personalized recommendation has received a lot of attention as a highly practical research topic. However, existing recommender systems provide the recommendations with a generic statement such as {"}Customers who bought this item also bought . . . {"}. Explainable recommendation, which makes a user aware of why such items are recommended, is in demand. The goal of our research is to make the users feel as if they are receiving recommendations from their friends. To this end, we formulate a new challenging problem called personalized reason generation for explainable recommendation for songs in conversation applications and propose a solution that generates a natural language explanation of the reason for recommending a song to that particular user. For example, if the user is a student, our method can generate an output such as {"}Campus radio plays this song at noon every day, and I think it sounds wonderful,{"} which the student may find easy to relate to. In the offline experiments, through manual assessments, the gain of our method is statistically significant on the relevance to songs and personalization to users comparing with baselines. Large-scale online experiments show that our method outperforms manually selected reasons by 8.2{\%} in terms of click-Through rate. Evaluation results indicate that our generated reasons are relevant to songs and personalized to users, and they attract users to click the recommendations.",
keywords = "Conversational recommendation, explainable recommendation, natural language generation, personalization, recommender system",
author = "Guoshuai Zhao and Hao Fu and Ruihua Song and Tetsuya Sakai and Zhongxia Chen and Xing Xie and Xueming Qian",
year = "2019",
month = "7",
day = "1",
doi = "10.1145/3337967",
language = "English",
volume = "10",
journal = "ACM Transactions on Intelligent Systems and Technology",
issn = "2157-6904",
publisher = "Association for Computing Machinery (ACM)",
number = "4",

}

TY - JOUR

T1 - Personalized Reason Generation for Explainable Song Recommendation

AU - Zhao, Guoshuai

AU - Fu, Hao

AU - Song, Ruihua

AU - Sakai, Tetsuya

AU - Chen, Zhongxia

AU - Xie, Xing

AU - Qian, Xueming

PY - 2019/7/1

Y1 - 2019/7/1

N2 - Personalized recommendation has received a lot of attention as a highly practical research topic. However, existing recommender systems provide the recommendations with a generic statement such as "Customers who bought this item also bought . . . ". Explainable recommendation, which makes a user aware of why such items are recommended, is in demand. The goal of our research is to make the users feel as if they are receiving recommendations from their friends. To this end, we formulate a new challenging problem called personalized reason generation for explainable recommendation for songs in conversation applications and propose a solution that generates a natural language explanation of the reason for recommending a song to that particular user. For example, if the user is a student, our method can generate an output such as "Campus radio plays this song at noon every day, and I think it sounds wonderful," which the student may find easy to relate to. In the offline experiments, through manual assessments, the gain of our method is statistically significant on the relevance to songs and personalization to users comparing with baselines. Large-scale online experiments show that our method outperforms manually selected reasons by 8.2% in terms of click-Through rate. Evaluation results indicate that our generated reasons are relevant to songs and personalized to users, and they attract users to click the recommendations.

AB - Personalized recommendation has received a lot of attention as a highly practical research topic. However, existing recommender systems provide the recommendations with a generic statement such as "Customers who bought this item also bought . . . ". Explainable recommendation, which makes a user aware of why such items are recommended, is in demand. The goal of our research is to make the users feel as if they are receiving recommendations from their friends. To this end, we formulate a new challenging problem called personalized reason generation for explainable recommendation for songs in conversation applications and propose a solution that generates a natural language explanation of the reason for recommending a song to that particular user. For example, if the user is a student, our method can generate an output such as "Campus radio plays this song at noon every day, and I think it sounds wonderful," which the student may find easy to relate to. In the offline experiments, through manual assessments, the gain of our method is statistically significant on the relevance to songs and personalization to users comparing with baselines. Large-scale online experiments show that our method outperforms manually selected reasons by 8.2% in terms of click-Through rate. Evaluation results indicate that our generated reasons are relevant to songs and personalized to users, and they attract users to click the recommendations.

KW - Conversational recommendation

KW - explainable recommendation

KW - natural language generation

KW - personalization

KW - recommender system

UR - http://www.scopus.com/inward/record.url?scp=85069505877&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069505877&partnerID=8YFLogxK

U2 - 10.1145/3337967

DO - 10.1145/3337967

M3 - Article

VL - 10

JO - ACM Transactions on Intelligent Systems and Technology

JF - ACM Transactions on Intelligent Systems and Technology

SN - 2157-6904

IS - 4

M1 - 41

ER -