Assessing Antidepressants Using Intelligent Data Monitoring and Mining of Online Fora

作者：Akay Altug; Dragomir Andrei; Erlandsson Bjorn Erik

来源：IEEE Journal of Biomedical and Health Informatics, 2016, 20(4): 977-986.

DOI：10.1109/JBHI.2016.2539972

• Data mining
• depression
• network analysis
• online fora
• semantic analysis
• social media
• user sentiment

Depression is a global health concern. Social networks allow the affected population to share their experiences. These experiences, when mined, extracted, and analyzed, can be converted into either warnings to recall drugs (dangerous side effects), or service improvement (interventions, treatment options) based on observations derived from user behavior in depression-related social networks. Our aim was to develop a weighted network model to represent user activity on social health networks. This enabled us to accurately represent user interactions by relying on the data's semantic content. Our three-step method uses the weighted network model to represent user's activity, and network clustering and module analysis to characterize user interactions and extract further knowledge from user's posts. The network's topological properties reflect user activity such as posts' general topic as well as timing, while weighted edges reflect the posts semantic content and similarities among posts. The result, a synthesis from word data frequency, statistical analysis of module content, and the modeled health network's properties, has allowed us to gain insight into consumer sentiment of antidepressants. This approach will allow all parties to participate in improving future health solutions of patients suffering from depression.

• 出版日期2016-7