Clause-level Relationship-aware Math Word Problems Solver

作     者：Chang-Yang Wu Xin Lin Zhen-Ya Huang Yu Yin Jia-Yu Liu Qi Liu Gang Zhou Chang-Yang Wu;Xin Lin;Zhen-Ya Huang;Yu Yin;Jia-Yu Liu;Qi Liu;Gang Zhou

作者机构：Anhui Province Key Laboratory of Big Data Analysis and ApplicationSchool of Data ScienceUniversity of Science and Technology of ChinaHefei 230026China Institute of Artificial IntelligenceHefei Comprehensive National Science CenterHefei 230088China Laboratory of Mathematical Engineering and Advanced ComputingInformation Engineering UniversityZhengzhou 450001China 

出 版 物：《Machine Intelligence Research》 (机器智能研究（英文版）)

年 卷 期：2022年第19卷第5期

页      面：425-438页

核心收录：

学科分类：08[工学] 081104[工学-模式识别与智能系统] 081203[工学-计算机应用技术] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by National Key Research and Development Program of China(No.2021YFF0901003) National Natural Science Foundation of China(Nos.61922073,U20A20229,and 62106244)。 

主　　题：Artificial intelligence(AI) artificial neural network(ANN) computational mathematics machine intelligence machine learning 

摘      要：Automatically solving math word problems,which involves comprehension,cognition,and reasoning,is a crucial issue in artificial intelligence research.Existing math word problem solvers mainly work on word-level relationship extraction and the generation of expression solutions while lacking consideration of the clause-level relationship.To this end,inspired by the theory of two levels of process in comprehension,we propose a novel clause-level relationship-aware math solver(CLRSolver)to mimic the process of human comprehension from lower level to higher level.Specifically,in the lower-level processes,we split problems into clauses according to their natural division and learn their semantics.In the higher-level processes,following human′s multi-view understanding of clause-level relationships,we first apply a CNN-based module to learn the dependency relationships between clauses from word relevance in a local view.Then,we propose two novel relationship-aware mechanisms to learn dependency relationships from the clause semantics in a global view.Next,we enhance the representation of clauses based on the learned clause-level dependency relationships.In expression generation,we develop a tree-based decoder to generate the mathematical expression.We conduct extensive experiments on two datasets,where the results demonstrate the superiority of our framework.