A Dual Framework And Algorithms For Targeted Online Data Delivery - 2011

ABSTRACT:

A variety of emerging on-line data delivery applications challenge existing techniques for information delivery to human users, applications, or middleware that are accessing info from multiple autonomous servers. In this paper, we tend to tend to develop a framework for formalizing and comparing pull-based solutions and gift dual optimization approaches. The initial approach, most sometimes used today, maximizes user utility below the strict setting of meeting a priori constraints on the usage of system resources. We have a tendency to tend to gift an alternative and a lot of versatile approach that maximizes user utility by satisfying all users. It will this whereas minimizing the usage of system resources. We have a tendency to have a tendency to debate the advantages of this latter approach and develop an adaptive monitoring solution Satisfy User Profiles (SUPs). Through formal analysis, we tend to see sufficient optimality conditions for SUP. Using real (RSS feeds) and artificial traces, we empirically analyze the behavior of SUP below varying conditions. Our experiments show that we tend to tend to can achieve a high degree of satisfaction of user utility when the estimations of SUP closely estimate the necessary event stream, and has the potential to save lots of a important quantity of system resources. We tend to more show that SUP will exploit feedback to boost user utility with solely a moderate increase in resource utilization.

Did you like this research project?

To get this research project Guidelines, Training and Code... Click Here