Solving General Arithmetic Word Problems

This paper presents a novel approach to au-tomatically solving arithmetic word problems. This is the first algorithmic approach that can handle arithmetic problems with multi-ple steps and operations, without depending on additional annotations or predefined tem-plates. We develop a theory for expression trees that can be used to represent and evalu-ate the target arithmetic expressions; we use it to uniquely decompose the target arithmetic problem to multiple classification problems; we then compose an expression tree, combin-ing these with world knowledge through a con-strained inference framework. Our classifiers gain from the use of quantity schemas that sup-ports better extraction of features. Experimen-tal results show that our method outperforms existing systems, achieving state of the art per-formance on benchmark datasets of arithmetic word problems.