[1]査 猛,叶 宁*,王汝传,等.基于胶囊网络模型的抑郁症预测研究[J].计算机技术与发展,2021,31(11):28-34.[doi:10. 3969 / j. issn. 1673-629X. 2021. 11. 005]
　ZHA Meng,YE Ning*,WANG Ru-chuan,et al.Research on Depression Prediction Based on Capsule Network[J].,2021,31(11):28-34.[doi:10. 3969 / j. issn. 1673-629X. 2021. 11. 005]
点击复制

基于胶囊网络模型的抑郁症预测研究()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
31
期数:
2021年11期
页码:
28-34
栏目:
大数据分析与挖掘
出版日期:
2021-11-10

文章信息/Info

Title:
Research on Depression Prediction Based on Capsule Network
文章编号:
1673-629X(2021)11-0028-07
作者:
査 猛1; 2 ; 叶 宁1; 2* ; 王汝传1; 2 ; 徐 康1; 2
1. 南京邮电大学 计算机学院、软件学院、网络空间安全学院,江苏 南京 210003;
2. 江苏省无线传感网高技术研究重点实验室,江苏 南京 210093
Author(s):
ZHA Meng1; 2 ; YE Ning1; 2* ; WANG Ru-chuan1; 2 ; XU Kang1; 2
1. School of Computer Science,Nanjing University of Posts and Telecommunications,Nanjing 210003,China;
2. Jiangsu High Technology Research Key Laboratory for Wireless Sensor Networks,Nanjing 210093,China
关键词:
微博; 抑郁症; 情绪词典; 胶囊网络; 文本分类
Keywords:
microblog; depression; emotional dictionary; capsule network; text classification
分类号:
TP391. 1
DOI:
10. 3969 / j. issn. 1673-629X. 2021. 11. 005
摘要:
抑郁症已经成为世界第四大疾病,并且具有高发病率,由于精神问题的诊断十分困难,抑郁症的治疗率很低。 随着互联网的发展, 社交媒体被越来越多的人使用,人们倾向于在社交媒体上表达自己的观点,这就为研究者创造了大量可以使用的数据, 但是抑郁症预测不等同于文本分类, 存在难以找到判定抑郁的统一标准,以及难以构建神经网络模型两个问题。 对此,提出了一种融合局部和整体特征的胶囊网络模型,该模型将情绪词典和胶囊网络进行融合,可以有效地发现微博用户中潜在的抑郁症患者。 模型分为局部特征选择和整体特征提取两部分,对于局部特征通过情绪词典进行选取,对于整体特征则使用胶囊网络模型进行学习,将两部分特征进行融合,得到该微博用户的抑郁症预测概率。 实验结果表明,提出的模型相较于用于文本分类的模型 KNN、DNN、TextCNN 和 BiRNN 等,在针对微博文本的抑郁症预测方面具有较高的准确率,能够有效地识别微博用户中潜在的抑郁症患者。
Abstract:
Depression has become the fourth largest disease in the world and has a high incidence rate. Due to the difficulty in diagnosing mental problems, the treatment rate of depression is low. Along with the development of the Internet,social media is being used more and more,people tend to express their views on social media, which will create a large amount of data that can be used for the researchers.However,depression prediction is not the same as text classification,there are two problems:it is difficult to find a unified standard to determine depression,and it is difficult to construct a neural network model. In this regard,a capsule network model combining local and overall features is proposed, which fuses the mood dictionary and capsule network,and can effectively find potential depressive patients among micro- blog users. The model is divided into two parts: local feature selection and overall feature extraction. For the local features,the mood dictionary is used for selection. For the overall features, the capsule network model is used for learning. The two parts of features are fused to obtain the prediction probability of the micro-blog user’s depression. The experiment shows that the proposed model is more accurate in predicting depression in micro-blog books and can effectively identify potential depressive patients among micro-blog users,compared with models used for text classification such as KNN,DNN, TextCNN and BiRNN.

相似文献/References:

[1]舒琰,向阳,张骐,等.基于PageRank的微博排名MapReduce算法研究[J].计算机技术与发展,2013,(02):73.
　SHU Yan,XIANG Yang,ZHANG Qi,et al.Research on MapReduce Algorithm of Micro Blog Ranking Based on PageRank[J].,2013,(11):73.
[2]潘明慧,牛耘. 基于多线索混合词典的微博情绪识别[J].计算机技术与发展,2014,24(09):28.
　PAN Ming-hui,NIU Yun. Emotion Recognition of Micro-blogs Based on a Hybrid Lexicon[J].,2014,24(11):28.
[3]苏小英[][],孟环建[]. 基于神经网络的微博情感分析[J].计算机技术与发展,2015,25(12):161.
　SU Xiao-ying[][],MENG Huan-jian[]. Sentiment Analysis of Micro-blog Based on Neural Networks[J].,2015,25(11):161.
[4]李梦洁,邵曦.基于文本属性的微博用户相似度研究[J].计算机技术与发展,2018,28(05):17.[doi:10.3969/j.issn.1673-629X.2018.05.005]
　LI Meng-jie,SHAO Xi. Research on Micro－blog User Similarity Based on Text Similarity[J].,2018,28(11):17.[doi:10.3969/j.issn.1673-629X.2018.05.005]
[5]房旋[],陈升波[],宫婧[][],等. 基于社交影响力的推荐算法[J].计算机技术与发展,2016,26(06):31.
　FANG Xuan[],CHEN Sheng-bo[],GONG Jing[][],et al. A Recommendation Algorithm Based on Social Influence[J].,2016,26(11):31.
[6]陈阳[],邵曦[],赵海博[]. 基于UR-LDA的微博主题挖掘[J].计算机技术与发展,2017,27(06):173.
　CHEN Yang[],SHAO Xi[],ZHAO Hai-bo[]. Microblog Topic Mining Based on UR-LDA[J].,2017,27(11):173.
[7]张凤娟,王濛,周刚.基于活动网络的微博用户影响力分析[J].计算机技术与发展,2018,28(09):162.[doi:10.3969/ j. issn.1673-629X.2018.09.033]
　ZHANG Feng-juan,WANG Meng,ZHOU Gang.Analysis of User Influence in Microblog Based on Activity Network[J].,2018,28(11):162.[doi:10.3969/ j. issn.1673-629X.2018.09.033]
[8]李 勇.一种改进的微博用户影响力分析算法[J].计算机技术与发展,2020,30(08):27.[doi:10. 3969 / j. issn. 1673-629X. 2020. 08. 005]
　LI Yong.An Improved Algorithm of Microblog User Influence Analysis[J].,2020,30(11):27.[doi:10. 3969 / j. issn. 1673-629X. 2020. 08. 005]
[9]盛丹丹.聚类算法在高校院所学生微博的应用研究[J].计算机技术与发展,2022,32(S2):47.[doi:10. 3969 / j. issn. 1673-629X. 2022. S2. 008]
　SHENG Dan-dan.Research on Cluster Algorithm in Institute of Geology and College Students’ Microblogging[J].,2022,32(11):47.[doi:10. 3969 / j. issn. 1673-629X. 2022. S2. 008]
[10]张石清,张星楠,赵小明.基于音视频信息的深度多模态抑郁症识别综述[J].计算机技术与发展,2023,33(07):1.[doi:10. 3969 / j. issn. 1673-629X. 2023. 07. 001]
　ZHANG Shi-qing,ZHANG Xing-nan,ZHAO Xiao-ming.A Survey of Deep Multimodal Depression Recognition Based on Audio-visual Cues[J].,2023,33(11):1.[doi:10. 3969 / j. issn. 1673-629X. 2023. 07. 001]

更新日期/Last Update: 2021-11-10