Glossary

Admin: profile with access to all eva functionalities. Analyst: profile that analyzes the bot performance. Answer: cell representing the bot response to a person.

API:  application programming interface, a communication protocol between different parts of a computer program.

Automated tests: the process of making sure that a bot answers an intent within expected parameters.

Bot: virtual agent capable of understanding human speech and responding accordingly.

Bulk training: massive import of intents to a bot knowledge base or examples/utterances to an intent.

Cell: the simplest structure in eva. It is an icon representing an action in a flow.

Channel: the place where the bot interacts with humans.

Cognitive engine: software that interprets natural (ie. Human) language. In eva, it is referenced only as NLP (Natural Language Processing)

Dashboard: the place where an editor, analyst or admin can check bot’s performance using metrics.

Dialog Manager: the place where an editor builds the knowledge base.

Editor: profile that creates and manages the bot knowledge base. Entity: cell representing the characteristics of an object in a flow.

eva: NTT DATA virtual agent - Virtual agents building and managing platform.

Evaluable Answer: answer that can be evaluated by the user.

Example/Utterance: a sentence that a human would say to a bot. It is the most important component of an intent.

Flow: string of cells that represents an interaction between a human and a bot. Dialog representation.

Input: cell representing a pause in a flow so the user can insert specific information to be sorted later by an entity.

Intent: cell representing what a user says to a bot.

Jump: the connection of cells that aren’t in the vicinity of each other.

Knowledge Base: the sum of all flows, intents, answers, services and trainings in a bot.

Metrics: measure of a data.

Not Expected Answer: the response of a bot to an intent that it does not recognize.

Not Expected Flow: an answer that works as a flow if the bot cannot recognize a user first intent.

Parameter: value added to configure software behavior.

Repository: the place where intents, answers, services, flows and trainings are stored.

Service: cell representing a connection of a flow to an external API.

Training: intent classification. The process that allows a bot to learn to interpret user input.

Transactional Answer: answer that has to be connected to an external API.

User: the person who is talking to the bot.

Welcome Message: The first message sent by a bot to a human in an interaction. It is an answer cell and works as a flow.

Workspace: the place where the editor will build a conversation flow.