High school weekly timetabling by evolutionary algorithms
Proceedings of the 1999 ACM symposium on Applied computing
Evolutionary Potential Timetables Optimization by Means of Genetic and Greedy Algorithms
ICIIS '99 Proceedings of the 1999 International Conference on Information Intelligence and Systems
ICTAI '03 Proceedings of the 15th IEEE International Conference on Tools with Artificial Intelligence
Towards multi-layer perceptron as an evaluator through randomly generated training patterns
AIKED'06 Proceedings of the 5th WSEAS International Conference on Artificial Intelligence, Knowledge Engineering and Data Bases
| Hi-index | 0.00 |
The school timetabling problem is a specific kind of timetabling problems. It is characterized by similar sets of subjects used among schools in different years, and by great extent of human factor involved. This particularity lets us to hope existing timetables to be useful information for actual timetabling process, and neural networks to be a suitable technique to assist it. This paper describes experiments on using neural networks as part of the fitness function of a GA-based school timetabling system, the model of what has been proposed by the author earlier. The experimental results show ability of neural networks to be applied for timetable evaluation, as well as reveal various side effects of using neural networks within GA-based school timetabling.