Research on multi-source data fusion and depth analysis method for fine management in universities

With the expansion of university scale and the improvement of management complexity, refined management has become an inevitable trend. This article focuses on the research of multi-source data fusion and depth analysis method for refined management of universities. It aims to build an algorithm model suitable for the fine management of universities by integrating multi-source data such as teaching, student management and logistics, and using multi-source data fusion technology, in-depth analysis technology and data preprocessing technology, including the prediction model of students' academic development and the prediction model of university resource demand. The experimental results show that the average root mean square error (RMSE), mean absolute error (MAE) and a coefficient of determination (R²) of the prediction model of students' academic development in different majors are 5.23, 4.12 and 0.85, respectively. The resource demand forecasting model of universities also shows good performance in forecasting classroom use area, frequency of experimental equipment use and total energy consumption. The average RMSE and MAE values of different campuses, universities and seasons are reasonable and R² is close to or more than 0.8. The research shows that the model can provide effective data support and decision-making reference for the fine management of universities, and help improve management efficiency and quality.