Neural network models for teaching multiplication table in primary school.

摘要

A novel idea of merging the profound and comprehensive modeling of the most low brain functions with education process is proposed. The main objective of the proposed methodology is to create a computer education system based on low level simulation of the pupil learning process. Comprehensive models of the human memory are possible only for the most fundamental processes of memorizing. Teaching mathematical facts in primary school is one of strict examples. A detailed model of the memory allows creating a computer educational system for teaching the multiplication table. The system stores pupil's answers in a neural associative memory and tunes the neural model to simulate the pupil's behavior. The computer model helps to select the most promising tasks guiding the pupil to learn educational materials in the best way and making the learning process more effective. The first tests in real school teaching process show the potential of the concept and open doors for further study of the approach. Additionally, the results of computer model application to the teaching process can be used as an immense data bank for testing and adjusting known models of the human memory.