Transductive Inference for Text Classification using Support Vector Machines
ICML '99 Proceedings of the Sixteenth International Conference on Machine Learning
Support vector machine active learning with applications to text classification
The Journal of Machine Learning Research
Kernel methods for relation extraction
The Journal of Machine Learning Research
Shortest-Path Kernels on Graphs
ICDM '05 Proceedings of the Fifth IEEE International Conference on Data Mining
A shortest path dependency kernel for relation extraction
HLT '05 Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing
Kernel-based learning for biomedical relation extraction
Journal of the American Society for Information Science and Technology
Employing personal/impersonal views in supervised and semi-supervised sentiment classification
ACL '10 Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics
BioNLP '10 Proceedings of the 2010 Workshop on Biomedical Natural Language Processing
Identifying potential adverse effects using the web: A new approach to medical hypothesis generation
Journal of Biomedical Informatics
Towards large-scale twitter mining for drug-related adverse events
Proceedings of the 2012 international workshop on Smart health and wellbeing
Social media mining for drug safety signal detection
Proceedings of the 2012 international workshop on Smart health and wellbeing
| Hi-index | 0.00 |
Post-marketing drug surveillance is a critical component of drug safety. Drug regulatory agencies such as the U.S. Food and Drug Administration (FDA) rely on voluntary reports from health professionals and consumers contributed to its FDA Adverse Event Reporting System (FAERS) to identify adverse drug events (ADEs). However, it is widely known that FAERS underestimates the prevalence of certain adverse events. Popular patient social media sites such as DailyStrength and PatientsLikeMe provide new information sources from which patient-reported ADEs may be extracted. In this study, we propose an analytical framework for extracting patient-reported adverse drug events from online patient forums. We develop a novel approach --- the AZDrugMiner system --- based on statistical learning to extract ad-verse drug events in patient discussions and identify reports from patient experiences. We evaluate our system using a set of manually annotated forum posts which show promising performance. We also examine correlations and differences between patient ADE reports extracted by our system and reports from FAERS. We conclude that patient social media ADE reports can be extracted effectively using our proposed framework. Those patient reports can reflect unique perspectives in treatment and be used to improve patient care and drug safety.