Abstract
Adverse drug reaction is a serious health problem that has a crucial impact on patient's morbidity, mortality and quality of life. This paper investigates the problem of detecting adverse drug reactions of anti-epileptic drugs from patients' reviews in online health forums. To this end, a lexicon-based methodology is proposed and applied to a dataset of patients' reviews collected from two online health forums. Following this methodology, the dataset is preprocessed using natural language processing techniques and the adverse reactions of anti-epileptic drugs are extracted with the aid of Consumer Health Vocabulary and a lexicon of adverse reactions. After that, proportional reporting ratio is applied to quantify the correlation between each drug and all adverse reactions and then identify the lists of adverse reactions for each drug. The detected lists of adverse reactions are validated quantitatively against a widely known database of adverse drug reactions called Side Effect Resource. The validation results provide empirical evidence on the effectiveness of the proposed methodology for the detection the adverse reactions from online health forums.