Procedural generation of problems for elementary math education

Mathematics education plays an essential role in children’s development, and there aremany online applications aimed at supporting this process. However, manually creat-ing math problems with a variety of textual and visual content is very time-consumingand expensive. This article presents a generic approach for procedural generation ofmathematical problems, including their corresponding textual representations. The con-tent generation process consists of two phases: abstract math problem generation andtext generation. For the generation of abstract math problems, we propose a generictemplate-based method that operates across a variety of difficulty-levels and domains,including arithmetic, comparison, ordering, mathematical relationships, measurement,and geometry. Subsequently, we propose a multi-language adaptive textual content gen-eration pipeline to realize the generated abstract math problems into semantically co-herent text questions in natural language. A workflow time gain evaluation shows thatthe system yields an average time saving of 56%. Further, human expert evaluation ofthis approach indicates that the content it generates is sensible and solvable for primaryschool students.

@Article { XSB21,
  author       = "Xu, Yi and Smeets, Roger and Bidarra, Rafael",
  title        = "Procedural generation of problems for elementary math education",
  journal      = "International Journal of Serious Games",
  number       = "2",
  volume       = "8",
  pages        = "49 -- 65",
  month        = "jun",
  year         = "2021",
  doi          = "10.17083/ijsg.v8i2.396",
  note         = "https://doi.org/10.17083/ijsg.v8i2.396",
  keywords     = "Procedural content generation, math problem generation, mathematics education, online education",
  url          = "http://graphics.tudelft.nl/Publications-new/2021/XSB21"
}