Recognizing the discipline category of the abstract text is of great significance for automatic text recommendation and knowledge mining. Therefore, this study obtained the abstract text of social science and natural science in the Web of Science 2010-2020, and used the machine learning model SVM and deep learning model TextCNN and SCI-BERT models constructed a discipline classification model. It was found that the SCI-BERT model had the best performance. The precision, recall, and F1 were 86.54%, 86.89%, and 86.71%, respectively, and the F1 is 6.61% and 4.05% higher than SVM and TextCNN. The construction of this model can effectively identify the discipline categories of abstracts, and provide effective support for automatic indexing of subjects.

Corporate finance courses are increasingly adopting data-driven teaching methods. Modern corporate finance courses are focusing more on students' career development. Through simulation practice and career planning guidance, students are better prepared to face challenges in the workplace after graduation. Students need to learn how to utilize data analysis tools and techniques to extract useful information from large datasets and make more accurate decisions. Data-driven teaching is a significant innovation in current curriculum reforms. In recent years, with the development of technology and the emergence of financial innovation, corporate finance courses have been undergoing continuous changes and innovations. These courses have started to emphasize emerging areas such as digital finance, blockchain technology, and sustainable development. Taking the example of corporate finance, this paper integrates the demands of skill development in the era of digital finance, focusing on aspects like teaching methods, reform methodologies, practical experiments, feedback mechanisms, and data analysis.

In the great practice of long-term revolution, construction and reform, the red gene created and developed by the CPC is a noble belief and noble emotion that has been continuously precipitated and inherited in the blood and struggle of countless people throughout the country, and has been deeply rooted in the blood and soul of the people of the CPC. Against the backdrop of the continuous development of modern education, integrating the red gene into the daily ideological and political education work of college students requires a clear understanding of its practical significance, and establishing the basic principle of integration based on the red gene, further promoting reform and innovation in inheriting and promoting the red gene in universities, and comprehensively enhancing the ideological and political awareness of college students, Provide a strong talent force for China's socialist construction.

Nowadays, the scale of graduate education in our country has been growing, but the quality of graduate education has not been improved. Therefore, how to effectively improve the quality of postgraduate education has become the most concerned issue in the academic circle and universities, which directly highlights that the internal guarantee mechanism of postgraduate students to improve the quality of postgraduate education has become the focus of academic research, in which tutors are the main influencing factors of postgraduate education quality. The tutor plays a positive and dominant role in stimulating, demonstrating, modeling, guiding and infecting the postgraduate's behavior. This paper analyzes the existing problems in exerting the role of postgraduate tutors, and from the problems, puts forward the countermeasures and suggestions to exert and mobilize the initiative of tutors.

In this study, the authors propose a method that combines CNN and LSTM networks to recognize facial expressions. To handle illumination changes and preserve edge information in the image, the method uses two different preprocessing techniques. The preprocessed image is then fed into two independent CNN layers for feature extraction. The extracted features are then fused with an LSTM layer to capture the temporal dynamics of facial expressions. To evaluate the method's performance, the authors use the FER2013 dataset, which contains over 35,000 facial images with seven different expressions. To ensure a balanced distribution of the expressions in the training and testing sets, a mixing matrix is generated. The models in FER on the FER2013 dataset with an accuracy of 73.72%. The use of Focal loss, a variant of cross-entropy loss, improves the model's performance, especially in handling class imbalance. Overall, the proposed method demonstrates strong generalization ability and robustness to variations in illumination and facial expressions. It has the potential to be applied in various real-world applications such as emotion recognition in virtual assistants, driver monitoring systems, and mental health diagnosis.