Work

Asking the Right Questions to Nominate A Student as Gifted and Talented: A Machine Learning Approach

Year: 2020
Type: article
Abstract: In this study, it is aimed to reduce the number of questions from a 69-item scale, which is developed to nominate a student as gifted and talented by selecting the most effective questions. For this purpose, Recursive Feature Elimination and Chi-Square Filter feature selection methods are used. Also, it is aimed to find the best performing machine learning algorithm to nominate a student as gifted and talented in this study. To achieve this, analyses are performed with Random Forest Algorithm, C4.5 Decision Tree Algorithm, and Naive Bayes Classifier machine learning algorithms. As a result of the analyses; the 69-item scale was reduced to 20 questions by using Chi-Square Filter method, and then when Naive Bayes Classifier was applied to this new data set, the model nominated a student with 92% accuracy as gifted and talented. It is thought that the proposed model will save time in the nomination process and prevent the distraction of attention that can be caused by the high number of questions when teachers fill out the scale. Also, it is believed that more rational decisions will be made in the nomination process by working with data-based prediction models.
Language: English
Open Access status: diamond
OpenAlex ID: W3095190662
FWCI: 0.481
Cites: 67
Cited by: 11
Related to: 10