>> The original co-occurrence system has access to all the data, including that from sessions exposed to the new algorithm. This is obviously a bad A/B Test de

>> The original co-occurrence system has access to all the data, including that from sessions exposed to the new algorithm. This is obviously a bad A/B Test design. The original algorithm shouldn´t have access to the data generated by the new algorithm. Designing an adequate test for ML systems is often as hard as designing the ML systems itself. And this is my main concern with machine learning.