Abstract: The present invention provides a new source of information, linguistic models, to improve the accuracy of mathematical recognition. Specifically, the present invention is an extension of linguistic methods to the mathematical domain thereby providing recognition of the artificial language of mathematics in a way analogous to natural language recognition. Parse trees are the basic units of the mathematical language, and a linguistic model for mathematics is a method for assigning a linguistic score to each parse tree. The models are generally created by taking a large body of known text and counting the occurrence of various linguistic events such as word bigrams in that body. The raw counts are modified by smoothing and other algorithms before taking their place as probabilities in the model.

Abstract: Provided herein is a pen-based editing system for manipulating mathematical expressions. Through a largely gesture based, directly manipulative interface, the system allows a user to make conventional changes to expressions, such as copy and move, and also to work with the expressions in ways peculiar to the problem domain, including, for example, handling ambiguity, expression fragments and alternate recognitions. The system is a generalization of an online recognizer for mathematical expressions. The system uses the same basic recognition techniques as the online recognizer, however the input information available to the editor is more varied, including mixtures of known and unknown characters and positional relations.

Abstract: The present invention provides a new source of information, linguistic models, to improve the accuracy of mathematical recognition. Specifically, the present invention is an extension of linguistic methods to the mathematical domain thereby providing recognition of the artificial language of mathematics in a way analogous to natural language recognition. Parse trees are the basic units of the mathematical language, and a linguistic model for mathematics is a method for assigning a linguistic score to each parse tree. The models are generally created by taking a large body of known text and counting the occurrence of various linguistic events such as word bigrams in that body. The raw counts are modified by smoothing and other algorithms before taking their place as probabilities in the model.