Equation Parsing: Mapping Sentences to Grounded Equations

TL;DR

This paper introduces Equation Parsing, a method to extract and generate mathematical equations from sentences by identifying variables and relations, achieving 70% accuracy in equation correctness and 60% in variable mapping.

Contribution

The paper presents the novel task of Equation Parsing, an efficient algorithm for mapping sentences to grounded equations, and provides a new annotated dataset for evaluation.

Findings

Correct equations generated in 70% of cases

Accurate noun phrase to variable mapping in 60% of cases

Outperforms baseline methods significantly

Abstract

Identifying mathematical relations expressed in text is essential to understanding a broad range of natural language text from election reports, to financial news, to sport commentaries to mathematical word problems. This paper focuses on identifying and understanding mathematical relations described within a single sentence. We introduce the problem of Equation Parsing -- given a sentence, identify noun phrases which represent variables, and generate the mathematical equation expressing the relation described in the sentence. We introduce the notion of projective equation parsing and provide an efficient algorithm to parse text to projective equations. Our system makes use of a high precision lexicon of mathematical expressions and a pipeline of structured predictors, and generates correct equations in $70\%$ of the cases. In $60\%$ of the time, it also identifies the correct noun phrase $\rightarrow$ variables mapping, significantly outperforming baselines. We also release a new annotated dataset for task evaluation.