深度学习在学生实验课异常行为检测中的应用  

Application of deep learning in abnormal behavior detection of student in experimental class

作　　者：蒋荣萍[1] 仇小烨 周卫[1] JIANG Rongping;QIU Xiaoye;ZHOU Wei(School of Artificial Intelligence,Guangxi Minzu University,Nanning 530006,China;School of Computer Science and Engineering&School of Software,Guangxi Normal University,Guilin,Guangxi 541004,China)

机构地区：[1]广西民族大学人工智能学院,南宁530006 [2]广西师范大学计算机科学与工程学院/软件学院,广西桂林541004

出　　处：《计算机应用文摘》2023年第21期71-74,共4页Chinese Journal of Computer Application

基　　金：2021年度广西高校中青年教师科研基础能力提升项目:基于深度学习的学生实验课异常行为识别(2021KY0170)。

摘　　要：近年来,随着技术的不断发展,深度学习在人工智能领域取得了优秀的成果,目前已进入多场景应用阶段。高校实验课是学生获取专业知识并掌握专业技能的重要实践渠道,其中的实验课堂表现能反映学生的上课专注度,可作为学生实验成绩的重要参考因素之一,但部分教师难以对其进行精确的评估。文章将人工智能融入实验课堂的教学中,通过行为识别实时监测学生的课堂行为,可精确评估学生的表现并确保课堂监测的准确性。In recent years,with the continuous development of technology,deep learning has achieved excellent results in the field of artificial intelligence and has entered the stage of multi scenario application.Experimental classes in universities are an important practical channel for students to acquire professional knowledge and master professional skills.The performance of experimental classes can reflect students'concentration in class and serve as one of the important reference factors for students'experimental grades.However,some teachers find it difficult to accurately evaluate them.The article integrates artificial intelligence into experimental classroom teaching,and monitors students'classroom behavior in real-time through behavior recognition,which can accurately evaluate students'performance and ensure the accuracy of classroom monitoring.

关 键 词：深度学习 课堂表现 行为识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]