Towards a Unified Index Scheme for Mobile Data and Customer Profiles in a Location-Based Service Environment

The increase in the demand for applications dealing with moving objects can be seen in recent years. Leaders in the mobile phone industry expect more than 1 billion mobile devices by 2004. Furthermore, mobile phones and/or wireless PDAs are expected to evolve into wireless terminals that are GPS enabled. In addition to wireless computing devices, tracking of other moving objects such as boats, trucks, automobiles, airplanes, and soldiers is also of growing interest. With the great demand on mobile devices, mobile commerce became a gigantic market opportunity. Some researchers predict the global subscriber base for mobile location services to exceed 680 million users by the end of 2006. Among those, 50% are mobile subscribers. They will represent more than 70% of the mobile Internet users. The revenues from mobile services have reached $2 billion by the end of 2002 and expected to reach $18 billion by the end of 2006. These revenues are divided as follows: 31% in Western Europe, 22% in the United States, and 47% in Japan and the rest of the World. As such, the market for location-aware mobile applications such as mobile shopping, mobile advertising, mobile retailing and mobile online banking is very promising. A study by Durlacher [2] shows that mobile advertising will be the killer application with 23% of the market share. The needs of such applications go beyond tracking users’ locations, for example, they may additionally need to track user profiles and preferences in order to achieve mass personalization. This is because, to be effective, targeted advertising should not overwhelm the mobile consumers and must push information only to a certain segment of mobile consumers based on their preferences and profiles, and based on certain marketing criteria. Obviously, these consumers should be targeted only if they are in the location where the advertisement is applicable at the time of the offer. It is important to note here that user profile information may include both sensitive and non-sensitive attributes such as name, address, linguistic preference, age group, income level, marital status, education level, etc.