- Avrupa Bilim ve Teknoloji Dergisi
- Issue: 38
- Prediction of Metacognition Awareness of Middle School Students: Comparison of ANN, ANFIS and Statis...
Prediction of Metacognition Awareness of Middle School Students: Comparison of ANN, ANFIS and Statistical Techniques
Authors : Seda Göktepe, Sevda Göktepe Yildiz
Pages : 450-461
Doi:10.31590/ejosat.1144623
View : 16 | Download : 5
Publication Date : 2022-08-31
Article Type : Research
Abstract :Problem-solving skill is one of the most important skills that an individual should have today. Reflection can best be observed in the problem-solving process because reflective thinking occurs when a particular problem is perceived. Since reflective thinking features are related to the individual’s own thinking processes, it has the feature of being a predictive variable for metacognition. This study’s main goal is to create models that predict middle school students’ mathematical metacognition awareness through reflective thinking characteristics towards mathematical problem solving utilizing Artificial Neural Network (ANN), the Adaptive Neuro-Fuzzy Inference System (ANFIS) and statistical techniques. Academic achievement scores, cumulative grade point average (GPA), and reflective thinking characteristics of students towards mathematical problem solving were used as input parameters while constructing the ANN and ANFIS model, and mathematical metacognition awareness of students served as the only output parameter. In addition, the system was trained using 70% of the data to build the ANFIS model. Feed-forward backpropagation with the Levenberg-Marquardt learning algorithm was used to train the network for ANN model. Statistically, there is no significant difference between the students' actual metacognitive awareness scores and the predicted ANFIS and ANN metacognitive awareness scores. These findings showed that the created models performed successfully in predicting the mathematical metacognitive awareness of middle school students through their academic achievement (general and mathematics) and reflective thinking features for problem-solving. This study serves as an excellent example of how artificial intelligence can be used to anticipate certain educational traits of students. Different applications of artificial intelligence in the area of education can be obtained by varying the methodologies employed in the research.Keywords : ANN, ANFIS, Fuzzy Logic, Mathematical Metacognition Awareness, Reflective Thinking Skill, Problem Solving