With the gradual penetration of artificial intelligence technology into various fields of society, it has brought many deeper and broader impacts, gradually improving the status of artificial intelligence in talent cultivation and education to adapt to the current development of social intelligence technology. Therefore, as the core course of artificial intelligence education in universities, machine learning needs to deeply analyze and explore the main factors that affect its development, in order to better mobilize students' learning enthusiasm and teachers' educational innovation, enhance the teaching and learning effectiveness of the course, and maximize the exploration of the educational achievements of artificial intelligence.

Arabic rhetoric has traditionally relied on ancient texts and human interpretation for teaching purposes. The study investigates ChatGPT’s ability to analyze and interpret Arabic rhetorical devices, specifically examining its capacity to handle cultural and contextual elements in rhetorical analysis. Drawing on institutional implementation frameworks and recent educational technology research, this study examines policy considerations for Arabic rhetoric education in an AI-driven environment, with a particular focus on sustainable digital infrastructure development and systematic reforms needed to support AI integration. The study employed the comparative approach to analyze eight rhetorical examples, including metaphors (“Zaid is a lion”), similes (“Someone is a sea”), and metonymy (“A person full of ash”), then compare ChatGPT’s interpretations with traditional explanations from classical Arabic rhetoric texts, particularly “Dala’il al-I’jaaz” by al-Jurjani. The results demonstrate that ChatGPT can provide basic interpretations of simple rhetorical devices, but it struggles with understanding cultural contexts and multiple layers of meaning inherent in Arabic rhetoric. The findings indicate that AI tools, despite their potential for modernizing rhetoric education, currently serve best as supplementary teaching aids rather than replacements for traditional interpretative methods in Arabic rhetoric instruction.

With the development of globalization, the importance of multiculturalism in university English education is becoming more and more prominent. This paper discusses the problems related to the teaching innovation of university English education from the perspective of multiculturalism. Firstly, it introduces the concept and significance of multicultural education and emphasizes its importance to university English education. Secondly, it analyzes the problems existing in the current university English teaching, such as cultural prejudice and language barrier. Then, innovative strategies for multicultural teaching are proposed, including teacher training, textbook design, teaching activities and so on. Finally, the evaluation and future development of multicultural education are prospected.

The six core competencies of mathematics in vocational schools are becoming increasingly important in mathematics learning. The reverse teaching design of vocational school mathematics, which focuses on core competencies, precisely grasps the internal logic of knowledge from a holistic perspective, and designs it on a unit by unit basis. The design process is to infer the starting point from the endpoint. Therefore, how to use reverse thinking in teaching design research in vocational school mathematics teaching under the background of core competencies will be the main content of this article.

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.