Research on Quality Monitoring and Data Mining Methods for Graduate Education in Innovation Management

Abstract

With the continuous growth in the scale of postgraduate admissions, monitoring the quality of postgraduate education has become an urgent issue that needs to be addressed. To this end, this paper establishes an information platform for monitoring the quality of postgraduate education from three aspects: structural design, functional design, and permission design. Based on the data analysis of this platform, data mining methods are used to study the learning behaviour and learning effectiveness of postgraduates. Using an improved K-means algorithm, graduate students are categorised into four types, with knowledge-exploration-oriented and marginally passive-oriented types being the most common, accounting for 37.99% and 27.60%, respectively. Multivariate regression analysis is then used to construct a multiple linear regression equation linking graduate students' learning outcomes to their learning behaviour characteristics. The regression coefficients for the number of knowledge tests and the module completion rate are 0.259 and 0.217, respectively, making them the primary factors influencing graduate students' academic performance. Utilising big data to analyse and predict graduate students' learning behaviours and learning outcomes facilitates monitoring the quality of graduate education and comprehensively improving the management level of graduate education