Use data mining methods in quality measurement in the education systems
Our basic problem is rooted in the education systems, where they want to measure and evaluate the pedagogical work from year to year and watch for pedagogical developments. These measures can be used for the individuals to get informations on which field they need to improve and can be used for rewa...
Elmentve itt :
| Szerzők: | |
|---|---|
| Testületi szerző: | |
| Dokumentumtípus: | Könyv része |
| Megjelent: |
2018
|
| Sorozat: | Conference of PhD Students in Computer Science
11 |
| Kulcsszavak: | Adatbányászat, Számítástechnika |
| Online Access: | http://acta.bibl.u-szeged.hu/61789 |
| Tartalmi kivonat: | Our basic problem is rooted in the education systems, where they want to measure and evaluate the pedagogical work from year to year and watch for pedagogical developments. These measures can be used for the individuals to get informations on which field they need to improve and can be used for rewarding systems. These measurements can be achieved by different methods, in this case our method is the surveying in the classes,direct approach person to person surveys and demand and satisfaction measurements for every person. These surveys are precomposed with discussions about the needed attributes. We get real datas from 58 different schools from 2007-2016 , nearly 8000 educators. All these surveys-ratings and other datas get collected and processed to get in a usable form. The schools make a statistics with the collected datas every year, but the statistics has really person dependents and because of that these has a lots of distorsions. For example a human focused teacher can give bad points for a informatics teacher class, because they dont share a territorial interest.To get a workaround for these statistics "personal" dependencies, we use the pagerank algorithm with the Comparability graph[1]. With the comparability graph we could compare two attributes with each other not heavily depended on the people who fill the survey and we could make a new graph for every attributes. After that on those graphs we can use the pagerank algorithms to get the datas out that we want to examine and get further consequences. For these datas we should tell they are lead to a much more usable development curves about the educators qualities. Important thing is these datas dont suffer the distorsions of the statistic ones has. |
|---|---|
| Terjedelem/Fizikai jellemzők: | 148-150 |