Character-Level Convolutional Networks for Arithmetic Operator Character Recognition

Computer-assisted instruction systems have been broadly applied to help students solve math problems, One of the challenges of intelligent instruction systems is how to make the machine automatically understand and recognize arithmetic operator characters. In this paper, we study the problem of automatic recognition of arithmetic operator characters based on character-level convolutional networks. Our work consists of three parts. The first part is to generate training data and test data for our convolutional neural network. The second is to split the input data into character-level and convert each character to a vector by using word embedding technology. The last step is to train our convolutional network and analyze the accuracy of recognitions of our proposed convolutional network. The experimental results show that our proposed network can correctly classify the arithmetic operator characters.