Personalized mathematical word problem generation

Abstract

Word problems are an established technique for teaching mathematical modeling skills in K-12 ed-ucation. However, many students find word prob-lems unconnected to their lives, artificial, and un-interesting. Most students find them much more difficult than the corresponding symbolic represen-tations. To account for this phenomenon, an ideal pedagogy might involve an individually crafted progression of unique word problems that form a personalized plot. We propose a novel technique for automatic gen-eration of personalized word problems. In our system, word problems are generated from gen-eral specifications using answer-set programming (ASP). The specifications include tutor require-ments (properties of a mathematical model), and student requirements (personalization, characters, setting). Our system takes a logical encoding of the specification, synthesizes a word problem narrative and its mathematical model as a labeled logical plot graph, and realizes the problem in natural language. Human judges found our problems as solvable as the textbook problems, with a slightly more artifi-cial language. 1