ABSTRACT With the rapid advancement of digital technologies, intelligent employment and entrepreneurship service systems have demonstrated significant potential in enhancing user experience and operational efficiency. However, cybersecurity challenges have emerged as a critical bottleneck for their development. These platforms accumulate vast amounts of user data—including resumes, job preferences, and emotional intelligence metrics. Our research has developed an emotion recognition‐based data collection system that analyzes real‐time facial expressions, voice patterns, and text content to extract users' emotional states. By applying machine learning algorithms to analyze and mine this data, we established an emotion classification model integrated into the intelligent platform, enabling personalized recommendations and emotional support services. For cybersecurity, our approach combines data encryption, access control, and identity authentication with blockchain technology for secure storage of emotional data. Additionally, adversarial training and federated learning methods were employed to enhance model robustness and privacy protection. Experimental results show that the intelligent system effectively captures users' emotional fluctuations through emotion recognition and machine learning, significantly improving engagement and satisfaction. The application of encryption and blockchain technologies ensures data integrity, confidentiality, and availability while preventing leaks and tampering. The overall performance and security of the system have been substantially enhanced. Integrating emotion recognition technology with machine learning and implementing robust cybersecurity measures allows employment and entrepreneurship service systems to deliver more personalized, secure, and reliable user experiences.
Xuna Wang (Sun,) studied this question.