Manual rule creation using DiscoveryOne’s Taxonomy Editor is still a viable option for clients to use. The main difference between the two approaches is time spent. With Machine Learning, once the necessary documents have been collected and the program has been started, we need only wait for the result and evaluate performance. However, it will be difficult for a human to interpret the rules found and validate their relevance. Machine Learning builds its strength on the law of large numbers, being able to evaluate and search for rules on a scale not achievable by humans.