Investigating Computational Identity and Empowerment of The Students Studying Programming: A Text Mining Study

In this study, it is aimed to predict the data obtained from answers given by students who receive programming education open-ended questions with text mining algorithms. Thus, text-based on computational identity and empowement were analyzed performances of different algorithms compared. The participants research consisted 646 whose age range was between 12-20 received education. An electronic form consisting prepared collect opinions A total six have been about (3 questions) empowerment questions). process followed in analysis set. Analyzes made Python 3.8 program. performance Word2vec (W2v) Term Frequency-Inverse Document Frequency (TF-IDF) word representation methods five machine learning compared: (a) Logistic regression, (b) Decision tree, (c) Support Vector Machines, (d) Random Forest, (e) Neural Network. Regarding identity, highest prediction accuracy found artificial neural network (tf-idf) logistic regression These an rate 93% regarding identity. It determined that method reached (96%) empowement. Following method, random forest (tf-idf), support vector 94%. fact these values are above 90% indicates estimation sufficient.