Chatbots Glossary: Terminology you need to know

Los Chatbots are revolutionizing the world of communication and customer service. Companies have seen how chatbots can help them reduce costs and improve engagement with their consumers.

In recent years, new communication channels have appeared, if companies had to have one or more people to answer all the queries made by users, the cost would be very high. Los benefits that chatbots offer to companies there are many, for example, automate processes and improve their efficiency.

But so that companies can Get the most out of chatbots, it is necessary to know What is a chatbot and how is it created. But even more important is to know what we are talking about, if a company wants to implement a chatbot, it needs to master or know the terminology of the industry or of the engineers or programmers who will create and configure the blog.

To make communication less confusing, we have prepared a glossary with the most important terms related to the Chatbots sector.

Glossary with the basic Chatbot terms you need to know:

Agent

An agent is a set of intentions around a topic.

AI chatbot (Artificial Intelligence Chatbot)

These are chatbots that are built with artificial intelligence techniques. This type of chatbot allows open conversations and has semantic understanding integrations.

Algorithm

An algorithm is a specific set of mathematical steps or operational commands that are used to solve a problem. Algorithms are fundamental to AI (Artificial Intelligence) because they analyze and transform data. Each analytical task requires a unique algorithm

Application programming interface (API) - Application Programming Interface

An API is a formal specification of how a module of a software communicates or interacts with another. It is a set of computer commands, functions and protocols (or methods, in object-oriented programming) that allow developers to create specific programs for certain operating systems or to be used by other software as a layer of abstraction.

Autoresponder

An automatic reply is a message that is automatically triggered when a user sends their first message, or specific keyword, to the bot. Have you ever written to a company on Facebook Messenger and have they responded to you right away? Surely it was an automated response from a chatbot.

Bots

A bot is software that attempts to understand a person's needs through natural language. Un Bot is capable of carrying out specific tasks and tries to understand human behavior.

Broadcast (broadcast)

A broadcast is a message proactively sent to a user. It's not a response to user feedback. Also known as “subscription messaging”, a broadcast is the chatbot equivalent of a push message in a mobile application.

Channel (Channel)

Means by which chatbot conversations enter. For example: Facebook Messenger, Skype, Slack and SMS.

Chat Logs

Chat Logs are transcript files of online chat conversations and instant messaging. Many chat or instant messaging applications allow online chat conversations to be archived by the customer, while a subset of chat or instant messaging clients allow chat files to be saved on a server for future retrieval.

Chatbot

A chatbot is the evolution of a bot, it is a service that uses artificial intelligence with which people interact through the interface of a chat.

Cloud or Cloud Computing

Cloud computing, also known as cloud services, cloud computing, cloud computing, cloud of concepts or simply “the cloud”, is a paradigm that allows computing services to be offered over a network, which is usually the Internet.

Context

Contextual data is chatbot information related to specific conversations and can be of relative importance. The objective is to collect as much data as possible for further analysis

Conversational UI (CUI)

A user interface based on human speech or language. It's everything that surrounds words. It's the design of the chat interface, the buttons, the widget, the images, the message templates.

Conversational UX (User Experience)

It refers to the experience of interacting with the bot. For example: what words do you use, how long it takes, if it has bugs, etc. Learn more about the user experience in a post in which we analyze the relationship between chatbots & UX, why they are so popular and what challenges exist for their improvement.

Deep Learning

It is associated with machine learning and depends on algorithms inspired by the structure and function of the human brain. Some forward-thinking chatbot engineers are looking for ways to harness the power of deep learning to make chatbots more emotional.

Entities

An entity is a variable that complements or modifies intent. It's a piece of information that the chatbot needs to understand in order to give the right answer. For example, if a user writes “show me the chatbot articles that Novicell wrote today”, the entities are “today” and “chatbots”. It refers to a specific day and not to a general question.

Framework (Work Environment)

A structure that provides the modules and functionality to build a chatbot that also requires programming.

Handoff

When a Chatbot is unable to provide an adequate response to a user, it is essential to transfer the conversation to a real person, including the context of the conversation so that the user doesn't have to start over with the agent.

Human in the Loop

Chatbots learn by collecting and monitoring data from conversations. The Artificial Intelligence system applies what it learns from each conversation. Human in the loop is when a human has to intervene to check and ensure that the appropriate response is given.

Intent (Intent)

An intention refers to what the user's expression means. An intention is something that the chatbot must be able to respond to. A typical chatbot is based on a series of attempts, along with an understanding of how it should respond to them. For example, if a user writes “show me the articles on Novicell digital strategy”, the user's intention is to retrieve a list of articles on digital strategy from the Novicell company. Intentions are given a name, often a verb and a noun.

Interaction

An interaction is a verbal or written communication between a chatbot and a person.

Interface

A shared border where two or more parts of a computer system exchange information.

Machine learning (ML)

Machine learning, or ML, is the field of artificial intelligence that focuses on making machines learn. Machine Learning plays a fundamental role in the evolution of chatbots because it allows the system to learn from each conversation by analyzing large amounts of data.

Markdown

Markdown is a lightweight markup language with plain text format syntax.

Neural Network

A computer system that follows the model of the human brain and the nervous system.

NLU (Natural Language Understanding)

It is a subset of NLP that deals with better managing unstructured entries and shaping them so that the chatbot can understand them and respond to the user.

NLG (Natural Language Generation)

This is when chatbots write text, LNG processes convert structured data into text (the processes that will be defined within the NLG) and will convert it into structured data so that the user can understand it.

NLP (Natural Language Processing) - Natural Language Processing

NLP is a field of computer science, artificial intelligence and linguistics that studies the interactions between computers and human language. (It includes both understanding, processing, which can be an action and/or a response to the user).

Pilot

This is the development stage in which a limited group of users test the chatbot we are creating.

Proof-of-Concept (POC) - Proof of Concept-

Concept tests are a method that is carried out with the purpose of verifying that the chatbot is susceptible to being exploited in a useful way. Tests are used to explain how zero-day vulnerabilities (vulnerabilities whose exact operation is unknown) can be exploited.

Push/pull (Push and Pull)

Push and yank These are terms that are often used to differentiate chatbots from more common marketing channels, such as email. For example, if a user asks you a question and you answer it, this is called a “pull” message, while if you write or send a message to a user without being asked by them, it is considered a “Push” message.

Rail, Guard Rails

When a user asks a question that baffles the chatbot or is beyond the chatbot's reach, the conversation must be redirected to a human.

Response

Anything the bot says in response to the user's question.

Rule-based chatbot

The behavior of this type of chatbots is determined by the rules we indicate to you. At each step of the conversation, users must choose one of the options shown to them to determine the next step. These types of bots have quite a few options, quick answers, and minimal text input.

Structured Data

Highly organized information that can be searched when placed in a database.

Training (Training)

When we talk about Training, we refer to improving the Chatbot's ability to manage queries. The process of forming a chatbot can vary from company to company. It should be noted that the training never ends, a chatbot can always improve, understand queries better, faster and with more precision.

Unstructured Data

Information that does not fit a predefined model or data structure and lacks general consistency

User Experience (UX)

The design of the user experience is a priority. The implementation of a user-centered design methodology from the beginning of a chatbot ensures that the user is at the center of each process and that the user experience works perfectly for the user and not for the designer.

Utterance (Dialogue)

It's everything the user says. For example, if a user types “show me Novicell's Digital Marketing services”, the full sentence is the statement.

Virtual Agent

It is a chatbot that is used for the interactions of the customer service, usually through a person.

But if we had to choose the most important concepts for creating and training a chatbot, these would be the Entities (Entities), the Intentions (Intents) and the Dialogo (Utterance).

In this post we have tried to compile a series of definitions related to the world of Chatbots. This is an article that we will be expanding in the future with new information. We would like you to participate and help us modify and expand the different terms related to Chatbots.

How can we help you?

If you need more information, do not hesitate to contact us.