Erik Swahn

Learn More
A recurrent neural network, modified to handle highly incomplete training data is described. Unsupervised pattern recognition is demonstrated in the WHO database of adverse drug reactions. Comparison is made to a well established method, AutoClass, and the performances of both methods is investigated on simulated data. The neural network method performs(More)
The aim of this thesis is to propose computationally feasible statistical methods for improved exploratory analysis of the complicated data sets involved in post-marketing drug safety monitoring. We implement an extended hit-miss model for duplicate detection in the WHO drug safety database and demonstrate its effectiveness on real world data. We propose(More)
  • 1