Application of Expert System with Fuzzy Logic in Teachers‘ Performance Evaluation
(Download reference research paper)
- Goal of this research paper is to show the implementation of artificial intelligence in form of fuzzy logics and expert systems. With the help of teachers’ performance assessment, it is easily explained that how the fuzzy variables can be used to quantify the qualitative data.
- First of all the attributes table is available to represent the required attributes for performance assessment of a teacher. Then appropriate weights are allocated to each attribute. Now the degree of relationship for each attribute is assigned with the help of fuzzy variables. It is the important phase which can help to easily convert qualitative data to quantitative. Because “VH, H, M, L, VL, Null” are easy to understand and to rate against an attribute. Then to ensure the correctness of results, main attributes are aided with sub-factors. By this nice fragmentation of attributes and weights allocation, research paper achieved its goals.
- Experimental results of this research paper which are extracted from the followings
- Proficiency in teaching
- Personal interest in teaching
- Presentation & comm. skills
- Speaking style & body language etc.
as mentioned in Table IV are finally evaluated by fuzzy expert system. It is an interesting part that qualitative data measured in depth and then finally can be described in linguistic description to get a meaningful and easy output.
Three cases A, B & C are taken as input to the fuzzy expert system. And results are Outstanding for A and Excellent for B & C. Which is meaningful in performance assessment of teachers.
- Future work which is identified in this research paper is regarding the assessment of employees in universities and other govt. and private organization. In fact this fuzzy expert system model is well explained in this research paper which can be extended to assess the performance in many sectors.
Although this research paper is well explained but still few portions can be improved. Here are few suggestions to improve this research paper
- Where three cases are taken as input in fuzzy expert system, these cases can be described with a little background to relate with reality. So further inspired implementations will become easier and quickly else the perception of different people will influence the input & results because of ratings.
- Pictorial representation should be improved, as we learnt from OOP lectures, Although OOP concepts were much difficult as compare to “Knowledge Refinement” picture in this paper.
- Although it’s a research paper, not a tutorial but still if formulae is given and explained so a little calculation can be given to show that are the exactly those values being fit in formulae which are we understanding from formulae. As we can get better understanding like a statistical exercise or research paper.
- Important terms are lost in paragraphs once you moved a couple of pages down like fuzzy variables and fuzzy membership functions etc. If we lost something so we have to read paragraphs again. Important terms should have bigger font size or bold style.
Last updated: March 19, 2014