Neural Networks with prototypes to quantify the academic achievement drivers: : evidence from a European country
Since the 1950s, academic performance has been the focus of interest by researchers and policymakers. However, only recently have data science methods begun to be applied more systematically to this topic. This work uses data from national mathematics and Portuguese exams of the Portuguese populatio...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Online |
| Language: | Portuguese |
| Published: |
Universidade Estadual de Campinas
2023
|
| Subjects: | |
| Online Access: | https://econtents.sbu.unicamp.br/inpec/index.php/tsc/article/view/17394 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Since the 1950s, academic performance has been the focus of interest by researchers and policymakers. However, only recently have data science methods begun to be applied more systematically to this topic. This work uses data from national mathematics and Portuguese exams of the Portuguese population in the 2018/2019 school year to, through neural networks, evaluate and compare which factors affect the results of these exams and in what way.Furthermore, a new approach is presented to deal with the "black box" dilemma of more advanced data science methods. This approach involves creating a set of prototypes through Neural Networks and estimating how much each potential impacts academic performance. |
|---|---|
| ISSN: | 2318-8839 |
