Academia.eduAcademia.edu

Figure 2 – uploaded by George Vachtsevanos

See full PDFdownloadDownload figure
Fig. 2. FGS functional block diagram.

Figure 2 FGS functional block diagram.

Related Figures (6)

Fig. 1. Generation of the fuzzy set “passing-score.”
Fig. 1. Generation of the fuzzy set “passing-score.”
polls’ domain {0, --. , 10}. The consequents “n_ medium” through “p_medium” are defined on the displacement universe [-1.5dmax, 1-5dmax]. The parameter dmax grossly controls the sensitivity of FGS. The contribution from each rule is computed as a scaled version of its consequent. The arithmetic sum is used to combine the fired rules across the rulebase. Finally, (1) is used to compute a crisp displacement.
polls’ domain {0, --. , 10}. The consequents “n_ medium” through “p_medium” are defined on the displacement universe [-1.5dmax, 1-5dmax]. The parameter dmax grossly controls the sensitivity of FGS. The contribution from each rule is computed as a scaled version of its consequent. The arithmetic sum is used to combine the fired rules across the rulebase. Finally, (1) is used to compute a crisp displacement.
draw four crosses on a supplied discretized line representing the [0, 100] scale. These crosses are placed where they woulc define the grade-boundaries for a given course level (senior graduate, etc.), in the absence of any students’ informatior (scores, GPA’s, etc.). Fig. 3 shows typical a priori fuzzy grade: created in this fashion. The height of some of the fuzzy grades need not be unity. Obviously the A and F fuzzy grades are always normal, unless at least one faculty member consider: that a perfect 100 is not an A (or a zero is not an F).
draw four crosses on a supplied discretized line representing the [0, 100] scale. These crosses are placed where they woulc define the grade-boundaries for a given course level (senior graduate, etc.), in the absence of any students’ informatior (scores, GPA’s, etc.). Fig. 3 shows typical a priori fuzzy grade: created in this fashion. The height of some of the fuzzy grades need not be unity. Obviously the A and F fuzzy grades are always normal, unless at least one faculty member consider: that a perfect 100 is not an A (or a zero is not an F).
Fig. 7. Mark distributions determined by curve methods and FGS for Situation 2. Fig. 6. Professor Y’s performance fuzzy vectors.
Fig. 7. Mark distributions determined by curve methods and FGS for Situation 2. Fig. 6. Professor Y’s performance fuzzy vectors.
Related topics:
EngineeringEducationFuzzy set theory
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved