Özet:

In this study, it is aimed to classify the self-efficacy beliefs of information technology teachers regarding the use of block-based coding tools by machine learning and deep learning methods. As a data collection tool, the T-SECT scale, which consists of previously developed likert-type items, was adapted to Turkish and used. The data set consists of 190 examples and 39 attributes of information technology teachers. In order to avoid the problem of unbalanced data due to the small number of samples, the data was amplified using the SMOTE method and the number of samples was increased to 262. The data set was transferred to WEKA software and machine learning methods and Convolutional Neural Network (CNN) were used on it. For this purpose, classification success was calculated with J48, Random Forest (RF), K-Star, Multilayer-Perceptron (MLP), NaivesBayes, SMO and IBK methods. The machine learning methods with the highest classification success on the data set were found to be SMO, MLP, IBK, J48 and RF. In the classification made with CNN, more successful results were obtained than the machine learning methods. Self-efficacy beliefs of information technology teachers were successfully classified using CNN with an accuracy rate of 99.30%.

Anahtar Kelimeler: