Development of Hybrid Algorithm Model for Students’ Achievement Forecasting

suwannee thoobjeen

Abstract


In this research, development of hybrid algorithm model by combination of causal forecasting methods using genetic algorithm and neural network and fuzzy logic is proposed for students’ achievement forecasting. The students’ learning achievement data set between the years 2013-2015 was used. The author selected 1,320 students’ grade: 660 for training and 660 for testing. From the results, the developed model can predict the most possible values that compare to the real values of both training data and testing data. A fuzzy rules system with 4 rules was generated by hybrid algorithm and had RMSE = 0.024 in the training data. The resulted fuzzy rules system produced RMSE = 0.087 on the testing data.


Keywords


hybrid system, causal forecasting, genetic algorithm, fuzzy logic, neural network

Full Text:

PDF

Refbacks

  • There are currently no refbacks.


SCIENCE  AND  TECHNOLOGY  RMUTT  JOURNAL
คณะวิทยาศาสตร์และเทคโนโลยี
Faculty of Science and Technology
http://www.sci.rmutt.ac.th/stj