A Study of Social Web Data on Buprenorphine Abuse Using Semantic Web Technology

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): The non-medical use of pharmaceutical opioids has been identified as one of the fastest growing forms of drug abuse in the U.S. There is a critical need to enhance current epidemiological monitoring, early warning, and post-marketing surveillance systems by providing additional and more timely data. The World Wide Web has been identified as one of the "leading edge" data sources for detecting patterns and changes in drug use practices. Many websites provide a venue for individuals to freely share their own experiences, post questions, and offer comments about different drugs. Such User Generated Content (UGC) can be used as a very rich data source to study knowledge, attitudes, and behaviors related to illicit drugs. To harness the full potential of the Web for drug abuse research, the field needs to develop a highly automated way of accessing, extracting, and analyzing Web-based data related to illicit drug use. This exploratory R21 is a multi-principal investigator, collaborative effort between researchers at the Center for Interventions, Treatment and Addictions Research (CITAR) and the Center for Knowledge-Enabled Information Services and Science (Kno.e.sis) at Wright State University. The purpose of this Web-based study is to apply cutting-edge information processing techniques, such as the Semantic Web, Natural Language Processing, and Machine Learning, to qualitative and quantitative content analysis of user generated content to achieve the following aims: 1) Describe drug users' knowledge, attitudes, and behaviors related to the illicit use of Suboxone(R) (buprenorphine/naloxone) and Subutex(R) (buprenorphine); 2) Identify and describe temporal patterns of the illicit use of these drugs as reflected on web-based forums. To collect data, the study will use websites that allow for the free discussion of illicit drugs, contain information on illicit prescription drug use, and are accessible for public viewing. The study will generate new information about the practices of buprenorphine abuse and will contribute to the advancement of public health and substance abuse research by providing automatic coding and information extraction tools needed to handle rapidly growing Web-based data. Automated information extraction methods applied in this study will enhance current early warning and epidemiological surveillance systems and could advance qualitative and Web-based research methods in other areas of public health.