摘要
机器学习是人工智能专业的核心课,课程具有工程实践性强、算法复杂且不易掌握的特点.基于工程教育的OBE教育理念,依据机器学习项目开发流程,利用课程算法设计了新浪微博热点话题研究的实验教学案例.将微博文本向量化后,基于TF-IDF算法赋予特征项权重并构建特征词矩阵,然后利用LDA模型采用EM算法得到不同时间窗口的主题,结合可视化技术发现微博热点主题.开展该教学案例的实验教学实践表明,使用机器学习算法解决社会热点问题的实验案例,不但能够较好地培养学生的实际工程能力和工程素养,而且激发了学生对该门课程的自主学习热情,是一种有效的实验教学案例设计方法;该实验综合性强,对学生的数理功底也是一种挑战.
The machine learning is a professional core course for artificial intelligence majors,and the course has the characteristics of strong engineering practice and complicated algorithms,which leads difficulty in mastering the knowledge in the course.The experimental teaching case of microblog hot topics research is designed according to machine learning project development process based on the result-oriented education concept of engineering education.After vectorizing the microblog text,the TF-IDF algorithm is used to assign weights to feature terms and construct a feature word matrix,and then the LDA model is used to obtain topics in different time windows using the EM algorithm,which is combined with visualization technology to discover hot topics on microblog.Experimental teaching practice of conducting this teaching case shows that the experimental case of using machine learning algorithm to solve social hot topics can not only cultivate students'practical engineering ability and engineering literacy,but also stimulate students'enthusiasm for independent learning of the course,which is an effective experimental teaching case design method.The experiment is highly comprehensive,which requires students'high mathematical abilities.
作者
董婧
程波
陈静锐
DONG Jing;CHENG Bo;CHEN Jingrui(School of Information Engineering,Qujing Normal University,Qujing Yunnan 655011,China;School of Marxism,Qujing Normal University,Qujing Yunnan 655011,China)
出处
《曲靖师范学院学报》
2023年第3期103-109,共7页
Journal of Qujing Normal University
基金
云南省科技厅省校联合基金项目“基于泊松簇过程建模的异构蜂窝网络在复杂衰落信道下的性能研究”(202101BA070001-091)
云南省教育厅科学研究基金项目“新媒体时代涉民族因素的网络舆情研究”(2023J1033)。
