This thesis proposes a framework for automation of data feed management and control. The framework can be classified into two broad areas i) building an ontology for a feed system and ii) developing a test suite, that is, a set of statistical models to analyze and diagnose feeds and formalize the results to feedback into the ontology. This thesis concentrates more on the latter and will only outline the course of action required for the former. Detailed descriptions of the statistical models that are part of the test suite are given and their strengths and weaknesses discussed. Practical results of implementing the test suite on a data feed system are analyzed. Various open and unexplored areas, issues and concerns are also discussed.
If you have any questions please contact the ETD Team, libetd@njit.edu.