Novel approach for Data Mining of Social Media to Improve Health Care using Network-Based Modeling

Social media provide a platform with tools to share information, to debate health care policy and practice issues, to promote health behaviors, to engage with the public, and to educate and interact with patients. More recently the growth of online social networks and healthcare forums has led patients to voluntarily share information about their health, treatments, and drug use. In healthcare sector sentiment analysis gives relevant information to the consumer about a particular drug. We proposed framework to identify user communities (modules) and prospective users. For this work, we go through user opinion(positive or negative) as well as a side effect of drug present in that user comments. In the first stage, we use the LVQ algorithm to perform sentiment correlation between user posts positive or negative opinion on the drug. Further to identify influential user and to get consumer opinion we used network partition method based on optimizing stability quality measures. Our technique can make bigger studies into intelligently mining social media information for consumer opinion of diverse treatments and helps to get up-to-date and rapid feedback to treatment and also help healthcare industry to provide effective treatments of future remedies.