This Small Business Technology Transfer (STTR) Phase I research project will explore the feasibility of developing a mathematical expression recognition engine that will adapt to the way a particular user writes mathematical symbols and expressions in real-time. A database of handwritten symbols and mathematical expressions will be created that will be used to prime an initial, writer-independent mathematical expression recognizer. This initial recognizer will use a single mathematical expression from a particular user to find a subset of the database which most closely matches that user's handwriting in an effort to tune the recognizer before real-time adaptation occurs. In this Phase I work, this idea of adaptive real-time learning for mathematical expression recognition will be brought from the concept to the prototyping stage.
The proposed recognition system will be incorporated into a pen-based software application for creating mathematical sketches which gives users the ability to make dynamic illustrations for visualizing mathematics and physics concepts by combining handwritten mathematical expressions with free-form drawings. The potential impact of this work will be the realization of a software tool that both teachers and students can use to augment learning in mathematics and the physical sciences.
