- Necmettin Erbakan Üniversitesi Ereğli Eğitim Fakültesi Dergisi
- Vol: 5 Issue: 1
- Investigating computational identity and empowerment of the students studying programming: A text mi...
Investigating computational identity and empowerment of the students studying programming: A text mining study
Authors : Nilüfer ATMAN USLU, Aytuğ ONAN
Pages : 29-45
View : 38 | Download : 41
Publication Date : 2023-06-24
Article Type : Research Article
Abstract :This study aimed to predict the texts obtained from the answers given by the students receiving programming education to open-ended questions, with text mining algorithms. Thus, an attempt was made to analyze text-based data in research on computational identity and programming empowerment and to compare the performances of different algorithms. The participants of the study consisted of 646 students studying programming with age range varies between 12-20. An electronic form consisting of open-ended questions was prepared to collect the opinions of the students who received programming education. There are a total of six open-ended questions about computational identity (3 questions) and empowerment (3 questions). The text mining process was followed in the analysis of the data set. Analyzes were carried out in Python 3.8 program In this study, Word2vec (W2v) and Term Frequency-Inverse Document Frequency (TF-IDF) word representation methods were used. Five machine learning algorithms compared in this study: (a) Logistic regression, (b) Decision tree, (c) Support Vector Machines, (d) Random Forest, (e) Artificial Neural Network. Concerning computational identity, it was found that the highest estimation accuracy was in artificial neural network (tf-idf) and logistic regression (tf-idf) algorithm. These algorithms have an accucary rate of 93% regarding computational identity. When the text-data related to programming empowerment was analyzed, it was determined that the logistic regression (tf-idf) method reached the highest accuracy prediction rate (96%). Following this method, random forest (tf-idf), support vector machine (tf-idf) and artificial neural network (tf-idf) algorithms predicted with 94% accuracy. The fact that these obtained scores are above 90% can be interpreted as sufficient estimation performance.Keywords : Bilgi-işlemsel kimlik, Programlama, Yetkilendirme, Metin madenciliği