Extracting Depression Symptoms from Social Networks and Web Blogs via Text Mining

• Published in: Bioinformatics Research and Applications Vol. 10330; pp. 325 - 330
• Main Authors: Ma, Long, Wang, Zhibo, Zhang, Yanqing
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 01.01.2017; Springer International Publishing
• Series: Lecture Notes in Computer Science
• Subjects: Depression symptoms; NLP; Social media; Text mining; Word clustering; Word embedding
• ISBN: 3319595741; 9783319595740
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-319-59575-7_29

Accurate depression diagnosis is a very complex long-term research problem. The current conversation oriented depression diagnosis between a medical doctor and a person is not accurate due to the limited number of known symptoms. To discover more depression symptoms, our research work focuses on extracting entity related to depression from social media such as social networks and web blogs. There are two major advantages of applying text mining tools to new depression symptoms extraction. Firstly, people share their feelings and knowledge on social medias. Secondly, social media produce big volume of data that can be used for research purpose. In our research, we collect data from social media initially, pre-process and analyze the data, finally extract depression symptoms.