SYMPOSIUM MINISYMPOSIA

ABSTRACT

A probabilistic neural network has been developed to reduce situationa l biases in the student ratings of university faculty instruction The purpose of the neural network is to improve the assessment of teaching effectiveness for p ersonnel decisions and faculty enhancement. Toward these objectives, this paper reviews the development of the neural network and subsequent application through Web-based interaction.

Three years of evaluation data were accumulated, analyzed, and used to train a p robabilistic neural network (PNN). The PNN produces a categorized distribution o f expected rating scores derived from the course situational factors. Factors us ed as input variables to the neural network include class enrollment, percentage of male students in the class, average GPA, expected grade, student ranking, ti me of the course offering, department, and type of course. The output variable f or network training is the mean rating. A CGI application was then developed to interface the C source output code from the PNN trained network with a Web-based input form.

Getting More Information

To obtain more information about the meeting send e-mail to: fscc98@pax.st.usm.edu.

Return to PAX
Return to USM

USM
inquire@delphi.st.usm.edu
Updated
http://pax.st.usm.edu