A2G Slides

October 2, 2005


agent, ecommerce, mobile computing

Agents2Go Slides

With the proliferation of mobile computing, more and more people use a variety of mobile devices in their dally lives. Recent years have also seen a remarkable growth in Electronic Commerce. The merger of these concepts has resulted in the emergence of Mobile Electronic Commerce (M-Commerce). One of the most critical requirements for M-Commerce is the ability to discover services in a given context. An important component of a user's context is their current location. For example, a user's on arriving at a location that he/she has never visited before should be able to find a local cab service. Current mobile devices have well known inherent limitations like limited power supply, smaller user interface, limited computing power, limited bandwidth and storage space. These limitations necessitate the development of systems that provide mobile users with high quality, precise and context relevant information. It is important that these systems be highly scalable since the demand for service searches will increase in the future. A location dependent search utilizes a user's current geographical location to refine their search and provide access to locally available services. One of the challenges of location-based searches is determining the user's current location. Users are often uncertain, or even completely unaware, of their current geographical location making location based searching more difficult. An automated detection of the user's current location would be very helpful in eliminating this problem. Location dependent systems are naturally described and implemented as distributed systems. This also improves their fault tolerance and scalability. For instance service information can be grouped by location and managed by a server responsible for the specified geographical region. In such a decentralized scheme user requests are processed at the local server and do not burden the rest of the system. This makes the system more efficient, responsive and scalable.


OWL Tweet

Past Projects

  1. Agents2Go