Personalized Diagnosis Analysis And Application Research Based On Learning Analytics Data

With the coming of the fourth Industrial Revolution,education has entered the era of intelligent teaching.As an important basic course in colleges and universities,the teaching quality of higher mathematics directly affects students’ subsequent professional learning,quality improvement and social skills enhancement.However,most colleges and universities still adopt the traditional teaching mode,which ignores the individual differences of students in teaching and fails to teach students according to their aptitude.Therefore,it is of great significance to use the intelligent means such as student portrait and personalized recommendation to diagnose the personality chemical situation of students’ higher mathematics learning and conduct personalized tutoring teaching for students according to the diagnosis results,which not only conforms to the development of The Times,but also can meet the needs of students.Aiming at this research topic,this paper mainly carried out the following two aspects of work:(1)The k-Means clustering algorithm is used to construct the learning situation portrait of the questionnaire survey data of students.Firstly,descriptive statistical analysis is made on the survey results,and it is found that students’ learning involvement will affect the difficulty of learning knowledge.Then deep neural network(DNN)was used to extract the recessive features of students’ learning situation data,and the recessive features of students’ learning engagement after dimensionality reduction were obtained.Finally,based on DNN extraction results,k-Means algorithm was used to construct clustering portraits of students,and based on the clustering results,the learning difficulties of each class of students were analyzed.The results show that students’ different levels of involvement correspond to different learning difficulties.Students with a higher level of engagement often face more challenging learning difficulties,while students with a lower level of engagement are more likely to encounter basic difficulties.(2)Construct a personalized learning resource system for students based on learning situation data.In order to facilitate the index and extraction of learning resources,the resources crawled by Python network were first classified and numbered,and then the knowledge graph was used to store advanced math exercises in triplet form.Then,by matrix multiplication of students’ recessive features and exercise knowledge graph,the personalized recommendation item database is obtained,and based on the recommendation database,personalized recommendation of advanced mathematics exercises for students is realized.At the same time,personalized video learning resources are recommended through the association between video learning resources and questionnaire logical jump questions.Finally,the satisfaction of the recommended students is investigated,which proves the effectiveness of the recommendation system.The results of this study demonstrate that personalized diagnosis based on learning situation data can effectively identify students’ learning characteristics,thereby achieving targeted teaching assistance.This method helps to improve students’ learning interest and autonomous learning ability,and enhance the learning effectiveness of higher mathematics.Therefore,this study provides a beneficial exploration and practice for personalized diagnosis in higher mathematics learning.