So, what we can do with the Platform?
General organization's settings
Adding a Telegram channel
Adding a Web channel
Creating a Script to use the Web Channel
Adding Viber channel
Adding a Facebook Channel
Adding a Twitter channel
How a SMS Integration Works
Contacts and Messages
Select a Message to Schedule for Later
Adding Media to the message
Static and Dynamic Groups
Flow editor and tools
Call Webhook: Making requests to external services
Split by Intent: Using Classifiers
Import and export flows
Expressions and Variables
Triggers and Campaigns
Adding a trigger
Tell a flow to ignore triggers and keywords
How to create a Campaign
Creating an Intelligence
Managing your team
Intents and Entities
A label is a keyword used to categorize important entities on texts.
See the examples below:
- I love soccer, it is a good sport.
- My father is a basketball coach.
- We played baseball on the last weekend.
On the phrases above, each highlighted word represents different kinds of sport, so we can infer that a good label to represent that group of words is a sport.
Why should I use this?
If it's necessary to give a higher level of detail on the message categorization, you should use this feature. For example, in a chatbot, if you want to send different messages to the user related to the kind of sport he/she likes, the first thing you would do is to identify the message intent, and after that you would like to verify which sport he/she is talking about, and to achieve this would be necessary to use entities and labels.
What is an entity?
An entity is a keyword element that will be marked with a label in a message body.
How can I use this?
When you are adding new samples to the training you must highlight all relevant entities on your training. See below:
- First thing you must do is to type the new sentence:
- After that you have to select the entity that you want to add to your training:
- Tap the white button below the phrase:
- On the white box that will appear, you must type the correct word that your bot should refer every time it identifies that word on the analyzed phrases. On the example, above we had "NYC" on the original text, but we know what it means, New York, in general, we must identify synonyms on the text and set a default word to each synonym, in this case 'New York'. Next step is to set a label to this word. Tap the button 'add the label' and type a keyword for that label:
- Then tap add. The new entity show receives a highlight. (You can add multiple entities on the same training sample).