Thesis Topic Details

Topic ID:
Context-aware mobile service recommendation
Raymond Wong
Research Area:
Information Retrieval, Mobile Computing, Services
Associated Staff
John Shepherd
Topic Details
R & D
Group Suitable:
Good programming skills; good Web programming or math knowledge is a plus.
Following the advent of large-scale wireless infrastructure some three decades ago, mobile phones have become the fastest growing technology in history and represent one of the largest known consumer markets. The devices have evolved from bulky brick-phones to slim flip-phones and now touch-sensitive smartphones. Today there are some 2.5bn mobile users with 3.3bn subscriptions among them, representing 40% of the population.

Use is widespread in developed countries with the US having 86% penetration and most of Western Europe boasting over 100% penetration due to multiple subscriptions. But the market has become saturated because those who want a phone already own a phone. This means that service providers compete for a share of existing users. They attract customers by offering a more attractive handset or more attractive price, and sustained revenue comes from subscriptions. A big money-maker is Value-Added Services, an industry term encompassing SMS, EMS, MMS, MIM, voicemail, email, web browsing, mobile TV, downloadable content (games, apps, video, music) and all services beyond end-to-end calls that 'add value' to use.

VAS make mobiles more functional, convenient and entertaining, but often incur a separate payment. Their price affects subscription rate and 3 in 4 users cite price as the critical deciding factor. Thus, lowering VAS prices should be a priority for service providers who want to increase market share.

One way to lower prices is to subsidize costs with advertising, but ads will need to be timely and accurate to be effective. Timeliness refers to ads being shown at opportune moments. Accuracy refers to ads being contextually relevant to the user activity, geographically relevant to all users in an area or personally relevant to the individual.

Existing mobile marketing approaches struggle with the timeliness and accuracy requirements. One of the current research directions is to discover context-awareness and use this context information to determine what ads will be of the user's interests.

This project will extend our previous work on mobile advertising and also on context-aware data mining. The final outcome will be a context-aware mobile advertising framework that addresses the issues and considers the changing user contexts. You will work with a PhD student in our group for this project (and optionally collaborate with one or two other international R&D groups).

Some reports of our previous work:
Past Student Reports
No Reports Available. Contact the supervisor for more information.

Check out all available reports in the CSE Thesis Report Library.

NOTE: only current CSE students can login to view and select reports to download.