Privacy aware obfuscation middleware for mobile jukebox recommender services

Ahmed Elmesiry, Dmitri Botvich

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

9 Citations (Scopus)

Abstract

Mobile Jukebox is a service offered by mobile operators to their clients, such that subscribers can buy or download anywhere, anytime fulllength music tracks over the 3G Mobile networks. Unlike some music download services, the subscribers can reuse the selected tracks on their music players or computers. As the amount of online music grows rapidly, Jukebox providers employ automatic recommender service as an important tool for music listeners to find music that they will appreciate. On one hand, Jukebox recommender service recommend music based on users’ musical tastes and listening habits which reduces the browsing time for searching new songs and album releases. On the other hand, users care about the privacy of their preferences and individuals’ behaviors regarding the usage of recommender service. This work presents our efforts to design an agent based middle-ware that enables the end-user to use Jukebox recommender services without revealing his sensitive profile information to that service or any third party involved in this process. Our solution relies on a distributed multi-agent architecture involving local agents running on the end-user mobile phone and two stage obfuscation process used to conceal the local profiles of end-users with similar preferences. The first stage is done locally at the end user side but the second stage is done at remote nodes that can be donated by multiple noncolluding end users that requested the recommendations or third parties mashup service. All the communications between participants are done through anonymised network to hide their network identity. In this paper, we also provide a mobile jukebox network scenario and experimentation results.

Original languageEnglish
Title of host publicationBuilding the e-World Ecosystem - 11th IFIP WG 6.11 Conference on e-Business, e-Services, and e-Society, I3E 2011, Revised Selected Papers
EditorsReima Suomi, Tomas Skersys, Rimantas Butleris, Lina Nemuraite
PublisherSpringer
Pages73-86
Number of pages14
ISBN (Print)978-3642272592
Publication statusPublished - 1 Jan 2011
Externally publishedYes
Event11th IFIP WG 6.11 Conference on e-Business, e-Services, and e-Society, I3E 2011 - Kaunas, Lithuania
Duration: 12 Oct 201114 Oct 2011

Publication series

NameIFIP Advances in Information and Communication Technology
Volume353
ISSN (Print)1868-4238

Conference

Conference11th IFIP WG 6.11 Conference on e-Business, e-Services, and e-Society, I3E 2011
Country/TerritoryLithuania
CityKaunas
Period12/10/1114/10/11

Keywords

  • Clustering
  • Mobile jukebox
  • Multi-agent
  • Privacy
  • Recommender service

Fingerprint

Dive into the research topics of 'Privacy aware obfuscation middleware for mobile jukebox recommender services'. Together they form a unique fingerprint.

Cite this