Gatus, Marivic M. 4 0

Development of a pseudo-expert system for academic evaluation / 6 6 Marivic M. Gatus. - - - xii, 139 pages 28 cm. - - - - - . - . - 0 . - . - 0 .

Thesis (M.A.) -- Pamantasan ng Lungsod ng Maynila, 2004.;A directed study presented to the faculty of Graduate School of Engineering in partial fulfillment of the requirements for the degree Master of Engineering Management (MEM) with specialization in Systems Management.





5



ABSTRACT: Expert system is a branch of Artificial Intelligence that makes extensive use of specialized knowledge to solve problems at the level of a human expert. It does not require the presence of the expert to do the actual process. Instead, the expert knowledge is acquired and codified into a real system. This directed study aims to develop a pseudo-expert system for academic evaluation. Its web-based design grants access to anyone connected to the Internet. Thus, the evaluation process is not constrained to a specified time and location where this process usually takes place. Furthermore, the system generates a list of possible subjects the students could enroll for the next semester giving due consideration to prerequisite subjects. Functionalities and features of the existing system are required for the development of the pseudo-expert system. Such functionalities and features are determined through the use of data gathered by interview, knowledge acquisition, document sampling and evaluation, and analyzed through the use of data flow diagram, visual table of contents, and input-process-output. The study evaluates and compares the result of the pseudo-expert system against the result of the manual evaluator based on the number of units computed. Statistical analyses, t-test and Sandler's A statistics, show that there is no significant difference between the two systems. Hence, the researcher concludes that the pseudo-expert system truly emulates the knowledge of the academic evaluator. Finally, the researcher recommends that further study be conducted with due consideration to the following: The validation of the CHED courses, all other courses taken by students who are already degree holders, and the inclusion of the scanned transcript of record as an actual entry to avoid encoding of data.













5







2 = =









2




2 --0------


6 --0-- 2 --------



0 2 --


--20------





--------20--


--------20--


----2

/ 2

/ 2

/

/