System for the recognition of online handwritten mathematical expressions

Most communication involving mathematical expressions is carried out over computer systems. Despite this fact, the entry of mathematical expressions for computer processing remains highly time-consuming and unintuitive. In this report the design and implementation of an automatic mathematical expression recognition system is presented with the goal of streamlining the human-computer-interface for entering mathematical expressions. The system functions by performing three operations: stroke grouping, symbol classification and symbol layout analysis. Stroke grouping is performed using a simple distance threshold and classification is performed using a support vector machine (SVM). Two methods are implemented and evaluated to perform the structural analysis: one based on the open-source DRACULAE parser and another on a custom adjacency matrix technique. The results show that the SVM has a 94.83% symbol classification accuracy while the DRACULAE parser and custom adjacency matrix technique have expression recovery rates of 80% and 43.3%, respectively. The high error rate of the custom technique is regarded as unacceptable and the DRACULAE parser is chosen as the preferred method. It is concluded that the system successfully achieves its goal of providing an intuitive interface for entering mathematical expressions.