antidepressants. They found that these forums offer a “valuable new data source.” And in 2010, the European Union funded the Psychonaut Web Mapping Project with the goal of developing a system to identify and categorize recreational drugs and new trends in drug use based on information on the internet.
Researchers who conducted the Twitter study had several key goals: Verifying that tweets can contain information about the misuse of prescription medication; annotating tweets that indicated misuse and creating a guideline for larger-scale, future research; and determining whether automatic annotation and classification systems for social media can be effective in monitoring medication misuse.
However, there were three main problems that researchers intend to address: Annotation needs to be done on a larger dataset in order to reduce ambiguities, a real-time monitoring tool needs to be implemented, and natural language processing techniques need to be utilized to identify non-personal tweets.
Despite the limitations, researchers stated that their automatic semantic analysis approach will have “significant importance for various toxicovigilance tasks” including determining the prevalence of abuse, analyzing patterns of nonmedical use of prescription medication, and determining how effective control measures are.
