摘要
针对传统方法仅从宏观层面对电影产业和影评进行计量统计和文字描述研究,无法有效挖掘高质量电影的主题和观众的评价,缺乏深层次的语义知识挖掘,提出一种基于LDA⁃BiLSTM模型和知识图谱的电影影评文本挖掘方法。首先,采集《你好,李焕英》电影的影评文本并进行预处理;其次,抽取影评的共现特征词,利用知识图谱挖掘电影质量及口碑相关特征词的关联关系,深入分析高质量电影的影响因素;最后,构建LDA⁃BiLSTM模型实现影评的情感分析,通过LDA模型提取影评的关键特征词,利用长短时记忆网络捕获长距离依赖关系,从而精准预测影评情感类别。实验结果表明,提出的方法能有效挖掘电影影评的情感特征词和关联关系,所提出LDA⁃BiLSTM模型的精确率、召回率、F1值和准确率依次为0.9839、0.9805、0.9822和0.9805,其结果优于其它机器学习和深度学习模型,为我国高质量电影挖掘提供学术思路,具有一定的研究价值。
Since the traditional method only conducts quantitative statistics and text description research on the film industry and reviews from the macro level,it cannot effectively mine the themes of highquality films and audience evaluations and needs deep semantic knowledge mining.This paper proposes a model based on LDABiLSTM and the knowledge graph for text mining of movie reviews.Firstly,this paper collects the film review text of“Hi,Mom”and preprocesses it.Secondly,it extracts the cooccurrence feature words of film reviews,uses the knowledge graph to mine the relationship between film quality feature words,and profoundly analyzes highquality movies.Finally,it builds the LDABiLSTM model to realize the emotional analysis of film reviews,extracts the critical feature words of film reviews through the LDA model,and uses the longshortterm memory network to capture longdistance dependencies to accurately predict the emotional category of film reviews.The experimental results show that the method proposed in this paper can effectively mine the emotional feature words and association relationships of movie reviews.The precision,recall,F1score,and accuracy of the proposed LDABiLSTM model are 0.9839,0.9805,0.9822,and 0.9805,which are superior to other machine learning and deep learning models.In short,this paper can provide academic ideas for mining our highquality movies.
作者
杨秀璋
武帅
廖文婧
项美玉
于小民
周既松
赵小明
Yang Xiuzhang;Wu Shuai;Liao Wenjing;Xiang Meiyu;Yu Xiaomin;Zhou Jisong;Zhao Xiaoming(School of Information,Guizhou University of Finance and Economics,Guiyang 550025,China;School of Information Management,Nanjing Agricultural University,Nanjing 211800,China;Guiyang School of Big Data and Finance,School of Big Data Application and Economics,Guizhou University of Finance and Economics,Guiyang 550025,China;Key Laboratory of Economics System Simulation of Guizhou,Guizhou University of Finance and Economics,Guiyang 550027,China)
出处
《现代计算机》
2023年第8期12-19,共8页
Modern Computer
基金
贵州省科技计划项目(黔科合基础〔2019〕1041、黔科合基础〔2020〕1Y279)
贵州省教育厅青年科技人才成长项目(黔教合KY字〔2021〕135)
贵州财经大学2021年度校级项目(2021KYQN03)。
