ABSTRACT

Objective: After a period of theoretical investigation and practical implementation in the realm of political and ideological education, an optimal educational system has been established. The primary objective of political and ideological education in higher education institutions is to facilitate the cultivation of a constructive worldview and the reinforcement of students' spiritual underpinnings. However, there is a noticeable disparity in the attention given to mental health awareness (MHE), as fewer courses are dedicated to this aspect. Extensive research has been conducted; however, it has been conducted in a manner that neglects the interdependence of the variables under investigation and the significance of their collaborative functioning. Proposed: The integration paradigm of political and ideological education and mental health education is addressed by the introduction of a Bayesian optimisation with bidirectional long-term memory (BO-BLSTM). The BO-BLSTM model suggested in this study aims to specifically target the recognition and classification of a wide range of political sentiments. Methods: This study proposes the utilisation of a Bayesian optimisation with bidirectional long-term memory (BO-BLSTM) within the framework of integrating political and ideological education with mental health education. The objective of the BO-BLSTM model is to discern the diverse categories of political sentiments that are present. Conclusion: The application of the Bayesian Optimisation (BO) technique allows for the optimisation of hyperparameters in the Bidirectional Long Short-Term Memory (BLSTM) model. Through comprehensive experimentation, it has been established that the BO-BLSTM model exhibits superior performance compared to existing approaches.