In order to develop a efficient chatbot, and exploit all the functionalities of the platform, you should know the terminology most used within the world of chatbots:
Interactions are the messages or communications that are sent between the chatbot and the user. These communications can have different formats:
Connections help to connect interactions with each other. There are two types of connections:
1. Locals: Local connections are those that communicate one interaction with another through a dashed line. These in turn, can be of two types:
- Conditioned: We use conditioned connections, when we want the user to perform some specific action before directing him to another interaction. The most used conditions are:
The user responds to anything
From the user’s response you can extract an email / phone / money / other…
The user’s response is equal to / contain / start…
- Unconditioned: If we connect two interactions with an unconditioned connection, the first interaction will be sent and then the second, without waiting for the user to take any action.
2. Globals: Global connections are similar to conditioned local connections. The difference is that local connections only work when the interactions are connected to each other through lines while, in global connections, they can be activated at any point in the conversation.
For example, if a user thanks the bot at any point in the conversation, you can create an interaction that is activated when it is detected and returns a thank-you message.
The channels are the means where we can deploy chatbots. Currently, with Wannabot you can deploy on the Web, Facebook, Telegram, or integrate it into your own App.
APIs allow you to connect your Wannabot-developed chatbots with third party services, which will allow you to further increase your chatbot’s functionalities.
The error message is used when the chatbot is not able to understand the user’s response, that is, it does not find a connection that meets the condition.
Wannabot offers different levels of error messages. Let’s see what types there are:
The acronym stands for “Natural Language Processing”. It is a branch of Artificial Intelligence thanks to which chatbots are able to understand users’ messages.
When the user writes a message, the Artificial Intelligence system, the first thing they do is try to identify their “intention” or “meaning”.
For example, if the user is in a shopping chatbot, he could write that he wants to buy a t-shirt. Previously we will have identified different ways of writing that same action: (I want to buy a T-shirt, I want to buy a T-shirt, I want to buy a T-shirt …). All these phrases would belong to the same “intention” = to buy.
We use entities to disambiguate a dialogue. It is possible that the same intention has a different meaning, and we use the entities to know the correct meaning of prayer. Let’s look at an example:
Example sentence 1: I would like to drink a juice
Example sentence 2: I would like to drink a soda
If we select soda and juice as entities we can disambiguate the meaning of the phrase, and guide you through different dialogues.
Entities are a limited subset of options that can be identified by the service, disambiguing the intention.