Online Healthcare Information Adoption Assessment Using Text Mining Techniques

Online healthcare community discussion forums facilitate patients and healthcare seekers to exchange knowledge on health-related issues. Mining the enormous volume of the user generated contents in these forums may provide valuable information regarding the health. The contribution of the responses in par with the question asked needs to be analyzed to provide meaningful solutions. Among the answers given by the repliers the answer that contributes knowledge for the health seeker or patient needs to be identified by analyzing the contents of the replies. The knowledge contribution of the reply is measured using the metrics such as Quality of information, Emotional support, and Source Trustworthiness, Replier Competition and Recipient Involvement. A knowledge Contribution Model proposed in this research work aims to determine the answers that could be adopted from the set of answers provided for each question. The pregnancy data from health-care community, Medhelp, is taken-up for analysis. Using Knowledge contribution model, the most influential answer that may be adopted from the given answers is determined by analyzing contents of the answers using text mining techniques. The SVM-RBF Kernel classification algorithm is applied to categorize the answers as adopted and non-adopted from the evaluated metrics.