Overview
Last updated
Last updated
Communication has been a cognitive activity since the dawn of time. In the early days, conversations were limited to verbal and textual interactions between people.
All of these interactions were guided by emotion, context, and knowledge of previous conversations. With the introduction of computers, the interactions have now extended to machines, i.e., human-machine interactions.
Artificial intelligence is a branch of knowledge that deals with the development of intelligent computer systems, i.e., systems that exhibit characteristics that we associate with human behavioral intelligence: Language comprehension, learning, reasoning, problem solving, etc.
Natural Language Processing (NLP) is a subfield of Artificial Intelligence that deals with the application of methods that enable computers to learn, understand, and produce content in natural language.
Using artificial intelligence and NLP capabilities enabled virtual assistants to understand user utterances in natural language, infer the task from the user utterance, and extract the information needed to perform the task successfully.
AI and NLP-based chats and virtual voice assistants are the latest trend in technology and a must-have for all businesses in this generation.
A Virtual Agent acts as an intelligent assistant between people and digital systems. It replaces the traditional user interfaces of an application or website with a conversational user interface. This is a paradigm shift from a previous communication, which involved either typing syntax-specific commands or clicking icons.
Virtual agents are designed to converse with the user through a combination of natural language conversations. Responses can take the form of buttons, calendars, or other widgets that speed up the user's response time. AI-powered messaging solutions, or conversational virtual assistants, serve as a springboard to the future.
They communicate via intelligent virtual agents, corporate apps and websites, social media, and messenger platforms. Users can interact with such assistants via voice or text to retrieve information, complete tasks and conduct transactions. So what makes the Conversational Virtual Assistant so special? In a nutshell:
A conversational virtual agent is based on three concepts:
Intents, is the intent of what the user wants to say to the bot, i.e., what the user expects the chatbot to say when it says something.
Utterances: these are the sentences that the user says to the chatbot.
Entities: these are keywords associated with the intentions that determine the response of the chatbot, as they are necessary for the execution of the action identified by
The Conversational virtual agent's job is to detect the intent and entities necessary to carry a conversation from the user utterance.
The intelligence of virtual agents is not innate, but must be trained with the help of machine learning, big data and NLP technologies.
A virtual agent is intelligent when it knows the user's needs, understands the context, and responds to the user based on the user's needs, mood, etc. This intelligence gives VA the ability to handle any scenario of a conversation with ease.
To design an intelligent VA it is necessary:
Create a natural language model that allows understanding user queries by detecting intentions. The NLP model must be trained by defining intentions using example texts (utterances). Eva allows you to create language models by using the custom NLP engine, creating intentions, defining utterances, and training entities. In addition, you can also use models trained in external engines such as Watson, Luis, or Dialogflow.
Design dialog flows for the different use cases we want to cover with the virtual agent. A dialog flow is a specific conversation to solve a use case. The flow includes language understanding, internal logic, calls to external services, and response management. With eva, we will be able to design flows of great complexity using the dialog manager.
Create the responses. Responses need to be written and customized to the tone and experience you want to deliver with the bot. Also, the responses can be customized to the experience that each conversational channel can provide. eva allows you to create responses for multiple channels so that you can make the most of each channel's experience.
