AI helps you reading Science

AI generates interpretation videos

AI extracts and analyses the key points of the paper to generate videos automatically


pub
Go Generating

AI Traceability

AI parses the academic lineage of this thesis


Master Reading Tree
Generate MRT

AI Insight

AI extracts a summary of this paper


Weibo:
Our experimental results show that the previous sequence of ads/songs ma ers in deciding what the right time is for exploration versus exploitation

Towards Effective Exploration/Exploitation in Sequential Music Recommendation.

RecSys Posters, (2018)

Cited by: 0|Views3
EI
Full Text
Bibtex
Weibo

Abstract

Music streaming companies collectively serve billions of songs per day. Radio-based music services may intersperse audio advertisements among the songs as a means to generate revenue, much like traditional FM radio. Regardless of the monetization approach, the recommender system should decide when to play content that the listener is know...More

Code:

Data:

0
Introduction
  • Recommender systems (RS) have been deployed in numerous domains including music, movies, e-commerce and books.
  • The authors conducted an experiment on a music recommendation application and the results show that the previous sequence of events in a listener’s session is important in deciding whether the RS should provide subsequent exploratory types of content.
  • To nd the e ect of di erent sequences of songs and ads on the probability of a user switching the station a er listening to an exploratory song, the authors looked at one million sessions on mobile devices where the ad placement had been made completely at random.
Highlights
  • Recommender systems (RS) have been deployed in numerous domains including music, movies, e-commerce and books
  • We conducted an experiment on a music recommendation application and our results show that the previous sequence of events in a listener’s session is important in deciding whether the RS should provide subsequent exploratory types of content
  • As you can see, depending upon the previous sequence of songs and ads, the probability of a user switching the station when we show them an explore song is higher than the probability when we show an exploit song. is is true for all 8 di erent combinations of songs and ads
  • We investigated the impact of di erent ad/song sequences on listener behavior
  • Our experimental results show that the previous sequence of ads/songs ma ers in deciding what the right time is for exploration versus exploitation
  • We will launch an A/B experiment controlling for the placement of explore songs and see how di erent users behave when they observe di erent sequences of songs and ads
Results
  • Percent Increase of Station Change Explore versus Exploit Song, following Sequence
  • The authors calculated the probabilities of users changing the station when they are exposed to di erent sequences of ads and songs as follows: there are a total of 8 possible event combinations for a set of three items as shown in gure 1.
  • P(C | S ) is the probability of a user changing the station given the last played content is an explore song.
  • P(C | S) is the probability of a user changing the station when the last played content is an exploit song.
  • Figure 1 shows the percent increase of station changes a er playing an explore versus an exploit song when a user has observed the respective prior sequence of exploit songs and ads.
  • As you can see, depending upon the previous sequence of songs and ads, the probability of a user switching the station when the authors show them an explore song is higher than the probability when the authors show an exploit song.
  • The ASA sequence has the highest probability increase (+531.13%) of a user switching the station when given an explore song a er that sequence.
  • Di erent sequences of songs and ads have di erent e ects on station switching behavior and a recommender system should try to take these sequences into account when doing exploration and exploitation, as in the sequential music recommendation system.
  • The authors focused on the impact of exploring new song content for the listener given the previous set of ads and songs in the listener’s session.
Conclusion
  • The authors' experimental results show that the previous sequence of ads/songs ma ers in deciding what the right time is for exploration versus exploitation.
  • The authors will launch an A/B experiment controlling for the placement of explore songs and see how di erent users behave when they observe di erent sequences of songs and ads.
  • The authors will investigate more sophisticated o ine models, such as HMMs and RNNs in a reinforcement learning se ing that could learn superior personalized playlist sequencing. is work is a starting point for a larger project in which the authors aim to optimize the stream of recommendations of mixed types of content [1, 3, 4]
Reference
  • Himan Abdollahpouri, Robin Burke, and Mobasher Bamshad. 2017. Recommender systems as multi-stakeholder environments. In Proceedings of the 25th Conference on User Modeling, Adaptation and Personalization (UMAP2017). ACM.
    Google ScholarLocate open access versionFindings
  • Himan Abdollahpouri, Robin Burke, and Bamshad Mobasher. 2017. Controlling Popularity Bias in Learning-to-Rank Recommendation. In Proceedings of the Eleventh ACM Conference on Recommender Systems. ACM, 42-46
    Google ScholarLocate open access versionFindings
  • Himan Abdollahpouri and Steve Essinger. 2017. Multiple stakeholders in a music recommender system. In 1st International Workshop on Value-Aware and Multistakeholder Recommendation at RecSys 2017
    Google ScholarLocate open access versionFindings
  • Robin Burke and Himan Abdollahpouri. 2017. Pa erns of Multistakeholder Recommendation. In 1st International Workshop on Value-Aware and Multistakeholder Recommendation at RecSys 2017.
    Google ScholarLocate open access versionFindings
  • Oscar Celma. 2016. e Exploit-Explore Dilemma in Music Recommendation. In Proceedings of the 10th ACM Conference on Recommender Systems. ACM, 377-377
    Google ScholarLocate open access versionFindings
  • Nofar Dali Betzalel, Bracha Shapira, and Lior Rokach. 2015. Please, not now!: A model for timing recommendations. In Proceedings of the 9th ACM Conference on Recommender Systems. ACM, 297-300.
    Google ScholarLocate open access versionFindings
  • Luiz Pizzato, Tomek Rej, omas Chung, Irena Koprinska, and Judy Kay. 2010. RECON: a reciprocal recommender for online dating. In Proceedings of the fourth ACM conference on Recommender systems. ACM, 207-214.
    Google ScholarLocate open access versionFindings
  • Paul Resnick, R Kelly Garre, Travis Kriplean, Sean A Munson, and Natalie Jomini Stroud. 2013. Bursting your ( lter) bubble: strategies for promoting diverse exposure. In Proceedings of the 2013 conference on Computer supported cooperative work companion. ACM, 95-100.
    Google ScholarLocate open access versionFindings
  • Hastagiri P Vanchinathan, Isidor Nikolic, Fabio De Bona, and Andreas Krause. 2014. Explore-exploit in top-n recommender systems via gaussian processes. In Proceedings of the 8th ACM Conference on Recommender systems. ACM, 225-232.
    Google ScholarLocate open access versionFindings
  • Xinxi Wang, Yi Wang, David Hsu, and Ye Wang. 2014. Exploration in interactive personalized music recommendation: a reinforcement learning approach. ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM) 11, 1 (2014), 7.
    Google ScholarLocate open access versionFindings
Author
Himan Abdollahpouri
Himan Abdollahpouri
Steve Essinger
Steve Essinger
Your rating :
0

 

Tags
Comments
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn
小科