Self-Organizing Neural Network Map for the Purpose of Visualizing the Concept Images of Students on Angles
Deniz Kaya · International Journal of Research in Education and Science · 2017
Thepurpose of the study is to perform a less-dimensional thorough visualizationprocess for the purpose of determining the images of the students on theconcept of angle. The Ward clustering analysis combined with Self-Organizing NeuralNetwork Map (SOM) has been used for the dimension process. The Conceptual Understanding Tool, whichconsisted of the open-ended question “write the first ten things you rememberwhen the term ‘angle’ is mentioned” to the study group, which consisted of 250seventh grade students. The analysis results showed that students mostlyexplained the concept of angle by associating it with mathematics and othersciences with the terms like “reflection and incidence angle, numbers, light,graphics, fraction, area, speed, algebraic expressions, energy, sound, fouroperations, viewpoint, electricity and natural numbers”. The students mostlyestablished relations with the static side of the angle. At the end of thestudy, the dataset obtained from the Conceptual Understanding Tool was used fortraining the SOM, and a structure that reveals the images of the students onangle has been recommended.