Dynamic Risk Assessment System for Classroom Teaching Process of Midwifery Major in Higher Vocational Colleges Based on Artificial Intelligence

Abstract

In the classroom teaching of midwifery majors in higher vocational colleges, the traditional dynamic risk assessment system lacks real-time performance, making it difficult to capture potential dynamic risks in the classroom in a timely manner. It also tends to focus on the basic level of knowledge mastery and learning attitude, and needs multi-dimensional assessment of students. This paper introduces artificial intelligence technology to build a more dynamic, comprehensive and real-time risk assessment system; it uses classroom management systems and sensor devices to collect students' behavior data, emotional fluctuations, learning progress and other multi-dimensional information in real time, and performs missing value filling, outlier detection and standardization processing. A dynamic risk assessment model was built based on XGBoost (eXtreme Gradient Boosting), and the model was trained to predict dynamic risks in the classroom in real time. The feature importance evaluation function of the XGBoost algorithm was used to select the features with the most predictive value automatically, and multiple comprehensive evaluation indicators were designed based on these features. Real-time data stream analysis was used to provide feedback to teachers on the classroom situation in combination with the dynamic risk assessment results, and teaching adjustment measures were recommended. The experimental results show that the number of classroom participation, eye tracking and classroom question-and-answer frequency in dynamic risk assessment are highly important, with importance values of 0.15, 0.14 and 0.13 respectively. In classroom risk identification, XGBoost has MSEs of 0.12, 0.18 and 0.14 respectively in identifying classroom risks such as mood swings, lagging behind in learning progress and decreased behavioral participation, and the recognition accuracy is 0.95, 0.87 and 0.90 respectively. The system response time is 9.5 seconds under 10,000 data streams, which has more efficient real-time processing capabilities; the experiment verifies the effectiveness of the research in this paper on the dynamic risk assessment system of the classroom teaching process of higher vocational midwifery majors.