Introduction
In the Validation section you can see the performance of the Model tested on the Ground Truth. The Model gets tested only on unseen data, due to the k-fold validation approach.
Usage
There are all Models of a project listed in the Validation dashboard as shown below. The Accuracy of the is shown in the last column after Model has finished the training process. To see the detailed validation of the Model can reached by clicking on theĀ ValidateĀ button.
The performance of a Model is shown in the Validate Model screen. In addition to common performance metrics and the confusion matrix, there is a detailed view of text extracts that have been classified correctly or incorrectly, as shown below.
In case the performance of the trained model stays below expectations, it can be helpful to try one of the following tips:
Try another template
Check which examples got miss classified and add more similar examples
In general, add more examples to the used Ground Truth
Try to build a successful proximity rule model first and use the bulk labeling feature to expand your Ground Truth set