The Optimization area in Teneo Studio Desktop provides various options to review a solution from different points of view in order to improve and optimize the conversational AI application; the section covers the following areas:
- Suggestions: the Suggestions cover different errors and potentially problematic issues detected automatically in a solution, the Suggestions page provide an overview of all errors and warnings as well as it provides ideas to how to resolve them.
- Intent Classification: the Classifier is the workbench for the machine learning model where developers can see how inputs from real logs are handled by the model, while the Class Performance allows to perform Cross Validation and run Test Data Evaluation on training and test data in order to check the performance of the intent model.
- Log Data: the pages here introduces the functionalities related to accessing, exploring, analyzing and manipulating log data produced by the live conversational AI application and introduces, among others, how to work with Log Data Sources, Query log data, and how to create and apply Augmenters
Teneo offers a number of functionalities which help and guide developers at improving the solution; this includes automatic detection of problematic issue and suggestion on how to solve these, proposals to enhance a machine learning model's balance or to assign inputs to a specific Class based on the models behavior, running Class Performance or reviewing Log Data from the live AI application to, for example, implement or improve Flows, answers and other aspects in the solution - all in order to enhance the solution and make it as good and effective as possible.