Predictive Analytics in C2P
The Approval Probability tool means we can predict the likelihood of whether or not proposed legislation will come into force, with an impressive *95% prediction accuracy. This allows you to focus on regulatory developments that are most significant to your business.
Using our data from C2P, the compliance knowledge management platform, we developed our own unique predictive analysis capability. Now when C2P users are alerted on new or changing regulations, we also alert them to the percentage chance this will come into force – the approval probability.
*Every day we train our algorithm for our ‘approval probability’ predictions. We collect all the regulations in C2P that were proposed and later either approved or rejected, and learn what parameters have contributed the most to their approval or rejection. To evaluate the accuracy of the prediction algorithm, we split regulations randomly into two sets – a training set and a test set. For each regulation from the test set we pretend that we do not know if it was approved or rejected. We use the algorithm discovered using the training set to predict if the regulation would be approved. We compare the predicted result with the actual fact, and confirm that they match in more than 95% of cases!