Learning From Human Decisions

In case a machine learning model is being used to make business decisions, the decisions can be marked to be reviewed by process SMEs before being used for further automation. In such scenarios, the inputs provided by the process SMEs are captured in the RPA transaction logs along with the ML Model inputs and outputs.

This logged data can be queried and can then be used for training and improving the accuracy of machine learning model. Following is a sample query that can be used to extract the above data for a process.