Development of teaching performance evaluation tool in higher education using artificial intelligent
How to improve the teaching quality in higher education, has become the current focus of the work of education. However in universities , classroom teaching is the main channel for the implementation of education. Its quality at a large extent reflects and determine the quality of education in...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2013
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/9464/1/CD7809.pdf http://umpir.ump.edu.my/id/eprint/9464/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| id |
my.ump.umpir.9464 |
|---|---|
| record_format |
eprints |
| institution |
Universiti Malaysia Pahang |
| building |
UMP Library |
| collection |
Institutional Repository |
| continent |
Asia |
| country |
Malaysia |
| content_provider |
Universiti Malaysia Pahang |
| content_source |
UMP Institutional Repository |
| url_provider |
http://umpir.ump.edu.my/ |
| language |
English |
| topic |
LB Theory and practice of education |
| spellingShingle |
LB Theory and practice of education Wu, Jing Development of teaching performance evaluation tool in higher education using artificial intelligent |
| description |
How to improve the teaching quality in higher education, has become the current focus of the
work of education. However in universities , classroom teaching is the main channel for the
implementation of education. Its quality at a large extent reflects and determine the quality of
education in colleges and universities. The evaluation is key to improve teaching quality. So how to
set up scientific justice evaluation of university classroom teaching quality system is very important
problem.
Evaluation system includes three basic parts: evaluation indicators and weights, sources of
information( the specific data of indicators) and the ways to deal with the information(models). We
briefly discussed the evaluation indicators, weights and the impact of that to teacher evaluation. We
mainly discussed and analyzed the way of data processing in the existent teacher evaluation system ,
and emphasized on utilizing a new way to solve the problems in traditional models.
Teacher evaluation is a highly non-linear relationship mixed with lots of qualitative and
quantitative. But in the existing teacher evaluation system, its is linear models that are used mostly
to deal with the information. First experts will set the specific indicators and the weights of each
indicator, and then gain the final result of evaluation through the weighted average of data. Though
the way is simple and easy to work, but the accuracy of the evaluation is not high, so generally it can
just evaluate the quantitative indicators, and it’s helpless to the qualitative or fuzzy indicators. In
addition, the weights of the indicators, in the evaluation system artificially made, which will cause a
big man-made effect on the evaluation process, that will result in some difference between the result
of the evaluation and the actual situation. Though the fuzzy and analytic way that came out recently
have solved problems on the qualitative and fuzzy indicators to a certain extent, but it has no much
improvement in solving the excessively dependence of evaluation on subjective factors, and in
reflecting the intrinsic relationship of indicators and the relationship between indicators and results,
and on the accuracy of the results, which make the evaluation lose the objectivity and scientific, and the reliability of the results is questionable.
This thesis attempts using the artificial neural networks method, appraises to the teaching
performance evaluation. We analyzed the characters on structure, content and using means of kinds
school’s teacher evaluation through online search and surveys on the spot, considering lots of factors
that can influence the results, putting forward improvement measures, establishing a BP net model to
deal with information of teacher evaluation, and optimizing the model processing by utilizing strong
functions of MATLAB toolbox. Finally 20 samples from a university in China, which are
representative in indicators, had a emulate exercise and validated test. The result was analyzed. The
data show that the model can objectively reflect the non-linear relationship between indicators and
results, the results are accurate, the precision is high, and the result has a good agreement with the
actual situation. According to the intrinsic relationship between indicator data and objectives based
on network, all weights of indicator come out automatically, so it can better solve the problems of
reliance of teacher evaluation on subjective factors, exclude the disturbance of personal factor, and
advance the reliability of evaluation. All the results show that applying BP network to teacher
evaluation is feasible and effective, and it will have a good prospect.
|
| format |
Thesis |
| author |
Wu, Jing |
| author_facet |
Wu, Jing |
| author_sort |
Wu, Jing |
| title |
Development of teaching performance evaluation tool in higher education using artificial intelligent |
| title_short |
Development of teaching performance evaluation tool in higher education using artificial intelligent |
| title_full |
Development of teaching performance evaluation tool in higher education using artificial intelligent |
| title_fullStr |
Development of teaching performance evaluation tool in higher education using artificial intelligent |
| title_full_unstemmed |
Development of teaching performance evaluation tool in higher education using artificial intelligent |
| title_sort |
development of teaching performance evaluation tool in higher education using artificial intelligent |
| publishDate |
2013 |
| url |
http://umpir.ump.edu.my/id/eprint/9464/1/CD7809.pdf http://umpir.ump.edu.my/id/eprint/9464/ |
| _version_ |
1755872506101628928 |
| spelling |
my.ump.umpir.94642023-01-12T04:36:23Z http://umpir.ump.edu.my/id/eprint/9464/ Development of teaching performance evaluation tool in higher education using artificial intelligent Wu, Jing LB Theory and practice of education How to improve the teaching quality in higher education, has become the current focus of the work of education. However in universities , classroom teaching is the main channel for the implementation of education. Its quality at a large extent reflects and determine the quality of education in colleges and universities. The evaluation is key to improve teaching quality. So how to set up scientific justice evaluation of university classroom teaching quality system is very important problem. Evaluation system includes three basic parts: evaluation indicators and weights, sources of information( the specific data of indicators) and the ways to deal with the information(models). We briefly discussed the evaluation indicators, weights and the impact of that to teacher evaluation. We mainly discussed and analyzed the way of data processing in the existent teacher evaluation system , and emphasized on utilizing a new way to solve the problems in traditional models. Teacher evaluation is a highly non-linear relationship mixed with lots of qualitative and quantitative. But in the existing teacher evaluation system, its is linear models that are used mostly to deal with the information. First experts will set the specific indicators and the weights of each indicator, and then gain the final result of evaluation through the weighted average of data. Though the way is simple and easy to work, but the accuracy of the evaluation is not high, so generally it can just evaluate the quantitative indicators, and it’s helpless to the qualitative or fuzzy indicators. In addition, the weights of the indicators, in the evaluation system artificially made, which will cause a big man-made effect on the evaluation process, that will result in some difference between the result of the evaluation and the actual situation. Though the fuzzy and analytic way that came out recently have solved problems on the qualitative and fuzzy indicators to a certain extent, but it has no much improvement in solving the excessively dependence of evaluation on subjective factors, and in reflecting the intrinsic relationship of indicators and the relationship between indicators and results, and on the accuracy of the results, which make the evaluation lose the objectivity and scientific, and the reliability of the results is questionable. This thesis attempts using the artificial neural networks method, appraises to the teaching performance evaluation. We analyzed the characters on structure, content and using means of kinds school’s teacher evaluation through online search and surveys on the spot, considering lots of factors that can influence the results, putting forward improvement measures, establishing a BP net model to deal with information of teacher evaluation, and optimizing the model processing by utilizing strong functions of MATLAB toolbox. Finally 20 samples from a university in China, which are representative in indicators, had a emulate exercise and validated test. The result was analyzed. The data show that the model can objectively reflect the non-linear relationship between indicators and results, the results are accurate, the precision is high, and the result has a good agreement with the actual situation. According to the intrinsic relationship between indicator data and objectives based on network, all weights of indicator come out automatically, so it can better solve the problems of reliance of teacher evaluation on subjective factors, exclude the disturbance of personal factor, and advance the reliability of evaluation. All the results show that applying BP network to teacher evaluation is feasible and effective, and it will have a good prospect. 2013-07 Thesis NonPeerReviewed application/pdf en http://umpir.ump.edu.my/id/eprint/9464/1/CD7809.pdf Wu, Jing (2013) Development of teaching performance evaluation tool in higher education using artificial intelligent. Masters thesis, Universiti Malaysia Pahang. |
| score |
13.211869 |