There is a limit to accurately recognizing ambiguous symbols with only the shape of handwritten mathematical expressions (HME). However, the accuracy of HME recognition can be further improved by using the learning contextual information when a learner inputs HME. This paper proposes a CIAR (Contextual Information Aided Recognition) system as a recognition method of HME using learning contextual information in the learning system. CIAR consists of HME recognition module, Learning contextual information generation module, and Learning contextual information recognition module. Each module recognizes HME, converts data obtained from learning contents and learning situations into information that can be used to improve the accuracy of HME recognition, and effectively utilizes the information to adjust the HME recognition results.