Effective Mobile Notification Recommendation Using Social Nature of Locations

Prasanta Saikia, James She

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Citations (Scopus)

Abstract

This paper proposes a robust method of location-based mobile notification recommendation. Sending the right notifications at the right locations at the right time can boost user experience of mobile applications. 1.8 million users' notification opening data is collected from a mobile news application for 2 months and analyzed. First, a way of grouping locations into by the similarities in their function or social nature is described for notification recommendation. The autocorrelations of these localities are then identified based on user preferences for news categories in them, and the presence of significant statistical noise is shown. The true latent correlations between localities are identified by performing robust correlation estimation using random matrix theory techniques. Finally, an effectiveness constrained notification optimization scheme, SLEECR, is presented that shows improvement in average notification opening rate of users by up to 62%, and decrease in average notification response time of users by up to 19%.

Original languageEnglish
Title of host publicationProceedings - 2017 IEEE 15th International Conference on Dependable, Autonomic and Secure Computing, 2017 IEEE 15th International Conference on Pervasive Intelligence and Computing, 2017 IEEE 3rd International Conference on Big Data Intelligence and Computing and 2017 IEEE Cyber Science and Technology Congress, DASC-PICom-DataCom-CyberSciTec 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1265-1270
Number of pages6
ISBN (Electronic)9781538619551
DOIs
Publication statusPublished - 2 Jul 2017
Externally publishedYes
Event15th IEEE International Conference on Dependable, Autonomic and Secure Computing, 2017 IEEE 15th International Conference on Pervasive Intelligence and Computing, 2017 IEEE 3rd International Conference on Big Data Intelligence and Computing and 2017 IEEE Cyber Science and Technology Congress, DASC-PICom-DataCom-CyberSciTec 2017 - Orlando, United States
Duration: 6 Nov 201711 Nov 2017

Publication series

NameProceedings - 2017 IEEE 15th International Conference on Dependable, Autonomic and Secure Computing, 2017 IEEE 15th International Conference on Pervasive Intelligence and Computing, 2017 IEEE 3rd International Conference on Big Data Intelligence and Computing and 2017 IEEE Cyber Science and Technology Congress, DASC-PICom-DataCom-CyberSciTec 2017
Volume2018-January

Conference

Conference15th IEEE International Conference on Dependable, Autonomic and Secure Computing, 2017 IEEE 15th International Conference on Pervasive Intelligence and Computing, 2017 IEEE 3rd International Conference on Big Data Intelligence and Computing and 2017 IEEE Cyber Science and Technology Congress, DASC-PICom-DataCom-CyberSciTec 2017
Country/TerritoryUnited States
CityOrlando
Period6/11/1711/11/17

Keywords

  • Notification
  • big data
  • mobile application
  • news

Fingerprint

Dive into the research topics of 'Effective Mobile Notification Recommendation Using Social Nature of Locations'. Together they form a unique fingerprint.

Cite this