Comparative assessment method between neural network & rubric

Abstract

This study aimed to design and develop to an E-portfolio assessment system on Project-based learning approach to teaching and learning oriented assessment. The researcher considered 60 pre-service senior teachers from two classes in university particularly on instruction media courses. The control group comprised 30 pre-service teachers who used E-portfolio system and analytic rubric assessment that was a coherent set of criteria for students' work. It included descriptions of levels of performance quality on the criteria. The experimental group comprised 30 pre-service teachers that served as who used E-portfolio system and neural network assessment. Neural Networks was a model of the work of human brain by using computer. It made computer as clever as the human learning, and trained to classify the data mining in E-portfolio. Experimental results indicate that the E-portfolio assessment system has no significant effect on pre-service teacher achievement, positive effect on self-learning. In addition, rubric assessment and neural network assessment of project-based learning achievements produced different results.