Predicting the industry income of the top 20 universities worldwide based on key performance indicators using a machine learning approach

In recent years, university-industry collaboration has emerged as a critical factor in the success of universities. This partnership can play a significant role in ensuring stability and securing additional resources for higher education. On the other hand, due to changes in the environment and the transformation of universities from third-generation to fourth-generation, their functions have also undergone change. Among these transformed functions, one of the most pivotal is industry income. In this context, the objective of this study is to forecast the indicators for industry income in the top 20 universities worldwide. This study draws upon the databases ( https://www.timeshighereducation.com ) and ( https://www.topuniversities.com ). Utilizing regression models, including Ridge, Least Absolute Shrinkage and Selection Operator (LASSO), and Elastic Net, as well as Artificial Neural Networks (ANN), we have analyzed the indicators of industry income within the studied universities over the past 12 years (2011 to 2022). The results reveal that, among the regression models, the LASSO model excels in predicting industry income. The ANN model, boasting a higher coefficient of determination (R2) and lower values of Root Mean Square Error (RMSE), outperforms penalized (PR) and linear (Ordinary Least Squares) models considerably, showcasing greater accuracy in this context. Overall, the findings underscore that research indicators, international outlook, and financial sustainability possess a greater predictive capability for augmenting industry income in the top 20 universities worldwide. This study adds to the readers’ knowledge about the industry income of universities as well as the factors that can help in increasing this type of income. Universities can pay attention to the results obtained in this study in order to diversify their incomes and also to strengthen the university-industry collaboration.