15 September 2025 in Education and Information Technologies
The importance of artificial intelligence (AI) literacy has grown significantly, and there is a rapidly increasing demand for AI professionals. However, AI faces a talent shortage, and the widening gender gap exacerbates this issue. Given the crucial role of secondary schools in nurturing students’ interest in AI and shaping their career paths, this study sought to investigate the link between secondary school students’ attitude toward AI learning, their AI literacy, and their interest in AI careers while examining possible gender differences. We used the AI Learning Attitude Survey, AI Literacy Survey, and AI Career Interest Survey to collect data from 622 secondary school students who were selected with a stratified sampling method. The survey data were then analyzed using structural equation modeling. The findings revealed that: (1) female students demonstrate lower levels of AI learning attitude, AI literacy, and AI career interest compared to their male counterparts, whereas no gender differences were found in the model of learning attitude-AI literacy-career interest; (2) positive AI learning attitude and AI literacy predict higher AI career interest, with learning attitude positively influencing AI literacy; (3) AI literacy significantly mediates the relationship between AI learning attitude and AI career interest. The findings highlight the necessity for K-12 schools to introduce both formal and informal AI education initiatives at an early stage. By fostering inclusive and supportive environments, schools can inspire all students, especially girls, to boost their AI literacy and consider career opportunities in this field. Such efforts will play a vital role in advancing diversity, equity, and inclusion within the AI sector.
14 April 2025 in IEEE Access
This paper addresses several challenges in classroom interaction analysis, including incomplete data collection, coarse-grained behavior classification, subjectivity in analysis, and the difficulty of aligning traditional methods with the evolving demands of modern education for precise and efficient teaching evaluations. We propose an automatic coding method for the Flanders Interaction Analysis (FIA) classroom observation record form (CORF) by integrating various artificial intelligence technologies to tackle these issues. Initially, audio is extracted from recorded classroom videos to isolate the speech of both teachers and students. Next, we employ the Paraformer model for speaker separation and speech recognition, converting the extracted audio into corresponding text and timestamp information for each speaker. Simultaneously, a pre-trained BERT module is fine-tuned on all spoken text. Finally, this fine-tuned BERT model is used to classify the text into ten distinct categories of interaction behavior, effectively generating the coding necessary for the CORF of FIA. Experimental results reveal that our proposed method achieves an accuracy of 94.63% in speaker separation, 95.32% in speech recognition, and 87.95% in agreement with manual annotations during speech behavior analysis. These findings indicate that our method is highly effective for automating the coding process of the CORF for the FIA system.
12 March 2020 in EURASIP Journal on Wireless Communications and Networking
The conventional method can make a reasonable analysis of common network attacks, but the reliability of the analysis is low under the virus propagation model. This paper proposes a new research method of computer network attack analysis based on the virus propagation model. Based on the relationship between the framework and daemon, the framework of the model for computer network attack analysis is set up, the attack analysis technology of computer network is determined, and the construction of the model for computer network attack analysis is completed. Computer attack objects and computer attack process are analyzed, and computer network attack analysis is carried out. Using the coverage test and the uncertainty test, the parameters of the reliability calculation variables are measured and the reliability calculation formula is replaced. It is concluded that the designed method of computer network attack analysis is 47.15% more reliable than the conventional analysis method, and is suitable for the network attack analysis under the virus propagation model.
18 March 2025 in IEEE Access
In China, students from ethnic regions often face significant challenges in mastering Mandarin, primarily due to the influence of their native languages. This struggle is further compounded by various factors, including inconsistent use of Mandarin by teachers, a lack of practical classroom resources, and a one-size-fits-all teaching approach, all of which hinder students’ spoken Mandarin proficiency and communication skills. To address these challenges, we propose an initial teaching activity framework that integrates advanced speech technologies across three key phases of the Mandarin teaching process: Motivating, enabling, and assessing, based on the Production-Oriented Approach theory. Then, we conducted three rounds of action research with 42 students to refine the proposed framework. At last, the effectiveness of the enhanced framework was evaluated using a univariate repeated ANOVA to analyze the results in Mandarin expression proficiency, complemented by comparative assessments of quantitative data and qualitative classroom observations collected throughout the research. The results indicate that the intelligent speech technology-assisted Mandarin teaching framework significantly enhances students’ Mandarin proficiency and effectively enables teachers to integrate technology into their teaching practices. This proposed framework has considerable potential for widespread application in initiatives to acquire Mandarin as a second language, equipping teachers with vital tools to improve their teaching strategies.
1 January 2026 in IEEE Access