SGO-DN-AF model for evaluating industry–academia collaboration in talent development

Abstract

Abstract Background Effective industry-academia collaboration (IAC) is critical for developing industry-ready talent in the age of digital transformation. Traditional assessment approaches for IAC frequently rely on linear or subjective analysis, which fails to reflect the multidimensional and nonlinear relationships between parameters such as research output, innovation capabilities, curriculum relevance, and job performance. Aim This research proposes a Squid Game Optimized Deep Network with Activation Function (SGO-DN-AF) for regression-based quantitative evaluation of IAC quality, integrating deep learning and metaheuristic optimization. Method Academic and industrial records were gathered from 3000 rows, including information on research projects, patents, artificial intelligence (AI)-based training programs, curriculum development, and student employability. SMOTE was utilized; min–max normalization and management of missing values were among the preprocessing processes. Principal Component Analysis (PCA) decreased redundancy, and an Autoencoder recovered latent feature representations. The Deep Neural Network (DNN) with ReLU activation function recorded nonlinear patterns, and its parameters were optimized with the SGO to improve convergence speed and avoid local minima, resulting in the SGO-DN-AF method. The SGO-DN-AF framework conducts multiclass classification to determine Industry-Academia Collaboration IAC quality levels, which can be classified into three categories: low, medium, and high. The target variable is the categorical IAC quality label derived from institutional and industrial performance indicators. Result Implemented in Python, Experimental results demonstrate that the proposed model outperforms baseline models, including standard DNN and ResNet architectures, Transformer models, and a conventional Neural Network-based model, achieving an MAE of 0.055. Curriculum innovation and AI integration were identified as key contributors to collaborative success. Conclusion The SGO-DN-AF method offers a strong, intelligent, and adaptable framework for quantitatively analyzing and improving IAC efficacy in talent development under Industry 4.0.