Title:
Depressive Emotion Detection and Behavior Analysis of Men Who Have Sex With Men via Social Media
Keywords:
depressive emotion detection 抑郁情绪检测
men who have sex with men 男男性接触者
behavior analysis
Blued
Abstract:
Background:A large amount of evidence has indicated an association between depression and HIV risk among men who have sex with men (MSM), but traditional questionnaire-based methods are limited in timely monitoring depressive emotions with large sample sizes. With the development of social media and machine learning techniques, MSM depression can be well monitored in an online and easy-to-use manner. Thereby, we adopt a machine learning algorithm for MSM depressive emotion detection and behavior analysis with online social networking data.
大量证据表明男男性接触者的抑郁与HIV风险之间存在关联，但是传统的基于问卷调查的方法在及时监测大样本抑郁情绪方面存在局限性。随着社交媒体和机器学习技术的发展，MSM抑郁可以通过在线和易于使用的方式得到很好的监控。因此，我们采用机器学习算法对MSM抑郁情绪进行检测和行为分析。
Methods:A large-scale MSM data set including 664,335 users and over 12 million posts was collected from the most popular MSM-oriented geosocial networking mobile application named Blued. Also, a non-MSM Benchmark data set from Twitter was used. After data preprocessing and feature extraction of these two data sets, a machine learning algorithm named XGBoost was adopted for detecting depressive emotions.
一个包括664335个用户和超过1200万个帖子的大规模MSM数据集是从最流行的面向MSM的地理社交网络移动应用程序Blued收集的。此外，还使用了来自Twitter的非MSM基准数据集。在对这两个数据集进行数据预处理和特征提取后，采用XGBoost机器学习算法检测抑郁情绪。
Results:The algorithm shows good performance in the Blued and Twitter data sets. And three extracted features significantly affecting the depressive emotion detection were found, including depressive words, LDA topic words, and post-time distribution. On the one hand, the MSM with depressive emotions published posts with more depressive words, negative words and positive words than the MSM without depressive emotions. On the other hand, in comparison with the non-MSM with depressive emotions, the MSM with depressive emotions showed more significant depressive symptoms, such as insomnia, depressive mood, and suicidal thoughts.
该算法在Blued和Twitter数据集中表现出良好的性能。提取出三个显著影响抑郁症检测的特征，包括抑郁词、LDA主题词和后时间分布。一方面，有抑郁情绪的男男性接触者发表的文章中抑郁词、消极词和积极词多于没有抑郁情绪的男男性接触者。另一方面，与有抑郁情绪的非男男性接触者相比，有抑郁情绪的男男性接触者表现出更显著的抑郁症状，如失眠、抑郁情绪、自杀念头等。
Conclusions:The online MSM depressive emotion detection using machine learning can provide a proper and easy-to-use way in real-world applications, which help identify high-risk individuals at the early stage of depression for further diagnosis.
基于机器学习的MSM抑郁情绪在线检测可以为现实世界的应用提供一种合适且易于使用的方法，这有助于识别抑郁症早期高危人群，以便进一步诊断。
Conclusion:
In summary, this is a new attempt to detect depressive emotions among MSM population using massive online social networking data with amachine learning algorithm. An effective and easy-to-use method is provided here for monitoring depressive emotions, which can help identify at-risk individuals in the early stage of depression for further clinical diagnosis. In addition, this is a novel analysis of the differences between MSM population and non-MSM population with or without depressive emotions. Automated depressive emotion screening via social media is a feasible and efficient measure for both the general population and hard-to-access populations. In the future, we expect to improve the representativeness of MSM population samples from online social media data and research the association between depression and stigma, and the sexual risk behaviors in MSM with or without HIV via online recruitment methods.
综上所述，这是一种利用大量在线社交网络数据和机器学习算法检测MSM人群抑郁情绪的新尝试。本文提供了一种简便有效的监测抑郁症的方法，有助于识别抑郁症早期的高危人群，以便进一步临床诊断。此外，本文还对有或无抑郁情绪的男男性接触者与非男男性接触者的差异进行了新的分析。通过社交媒体进行的抑郁症自动筛查对于一般人群和难以接触的人群来说都是一种可行和有效的方法。未来，我们期望通过在线社交媒体数据提高MSM人群样本的代表性，并研究抑郁与污名之间的关联性，以及通过网络招募方式对携带或不携带HIV的男男性接触者的性风险行为进行研究。
