首页|基于自然语言处理的学生评教情绪分析

基于自然语言处理的学生评教情绪分析

扫码查看
对学生评教信息中蕴含的情绪分析对于课堂教学的改进起着至关重要的作用,使用了"中文分词+to-ken+LSTM模型"的自然语言处理方式对学生评教信息进行了情绪分析。设置词表和停用词,对数据集进行中文分词。将得到的中文分词列表训练得出数字字典,将分词列表转换成数字列表,最后将数字列表转成空间向量形成数据集。建立LSTM模型,使用建立好的训练集进行训练,对训练后的模型进行评估,评估结果证明该模型是可靠的,对选取的典型的和复杂的数据进行预测,得出情绪分析结果。实验证明,该模式对于典型和复杂评教信息的分析结果均是正确的。
Emotion Analysis of Student Evaluation of Teaching Based on Natural Language Processing
The emotion analysis of students'teaching evaluation information plays a crucial role in the improvement of class-room teaching.This paper uses the natural language processing method of"Chinese word segmentation+token+LSTM model"to carry out the emotion analysis of students'teaching evaluation information.Set thesaurus and stop words,and divide the data set in-to Chinese words.The Chinese word segmentation list is trained to get a digital dictionary,the word segmentation list is converted into a number list,and the number list is converted into a spatial vector to form a data set.The LSTM model was established and trained with the established training set.The trained model was evaluated,and the evaluation results proved that the model was reli-able.The selected typical and complex data were predicted and the results of emotion analysis were obtained.The experimental re-sults show that the model is correct for the analysis of typical and complex teaching evaluation information.

natural language processingevaluation information of teachingemotion analysisChinese word segmentationLSTM model

高云、刘寰、周建慧、郭艳萍

展开 >

山西大同大学计算机与网络工程学院,山西大同 037009

大同市职业教育中心,山西大同 037000

自然语言处理 评教信息 情绪分析 中文分词 LSTM模型

山西省软科学研究计划

2019041023-5

2024

山西大同大学学报(自然科学版)
山西大同大学

山西大同大学学报(自然科学版)

影响因子:0.271
ISSN:1674-0874
年,卷(期):2024.40(5)