Intent
Last updated
Last updated
Intents are the “heart” of every virtual agent, refers to the goal the customer has in mind when typing in a question or comment. When designing a conversational flow you need to predict the user interactions and the virtual agent answers.
An intent cell will represent the users' interactions and will, in other words, an action, hence it's usually associated with a verb.
Since users can express the same will in different ways, the intention cell is composed of examples. The examples are all the possible ways the user can express the same will.
Examples of a purchase intention:
I want to buy
I want to acquire
I want to make a purchase
If you are a developer, check the intents table for more information.
With eva NLP, the Intention is the first act in creating a dialogue. And here are some tips for creating them according to Best Practices:
In the “Name” field, use _ (underline) or – (hyphen). Spaces are not allowed there.
Pay attention to the real desire of your user when you give the name of an Intent. For example, the Sentence/Utterance “I want a soup” means that, more than a desire to eat a soup, the user wants to make an order. So, you should name the Intent as make_order.
In the field “Add example/Utterance”, write some sentences that your user should express the Intents. For example, to the Intent make_order, you should write sentences such as “I wanna a soup, please”, “May I get a soup?”, “I’m starvy and I whant to order right now!” or even typing and linguistic error, such as “whanna a soup”.
Avoid similar examples in different intents. This might confuse the virtual agent when choosing an intent.
Not only check what your user says, but how your user says. People talk differently between channels. Pay attention to regional patterns, slang, social groups and writing proficiency.
Before giving the name of the Intent and register your Utterances (always remembering that Utterance is a sentence/phrase that indicates the Intent of an user), you click the button “Save”. And that’s it!
Very Important:
Be very careful when deleting Intents!
In any NLP, if you delete an Intent on the beginning of a flow, the flow will automatically be deleted as well.
When using NLP Watson, if you change the name of any Intent, even if it is an Intent used in the middle of a flow, it will automatically be deleted in eva. Important note: this is specifically when using Watson.
Also Very Important:
If you're using eva NLP (NTT DATA proprietary Natural Language Processing engine), you must register at least 5 Intents, each one with at least 5 examples/utterances to start your virtual agent.
You might have a lot of Intents or examples/utterances. To save the work of registering each one individually, eva allows you to upload those intents and examples/utterances in one file.
To import more than one Utterance/Intent, first you have to make sure that they are written following certain standards:
First, create a CSV file.
Then, edit the csv file separating Utterances to Intents with a comma, and do not use spaces, like the model below: Hi,Greeting or I want pizza,Requisition (in the first sentence, Hi is the Utterance and Greeting, the Intent. In the second, I want pizza, the Utterance and Requisition, the intent).
Never use diacritics (´`¨^).
Write only one sentence per line.
If your phrase has a comma, put it between a quotation mark, for example: “Hello, dear virtual agent, I want pizza”,Requisition or “Hello, virtual agent!”,Greeting.
If your intent has more than one example/utterance, you have to write each one, followed by the intent, like the model below:
Example/utterance1,intent1 Example/utterance2, intent1
Check the examples below:
I need computer,place_order
I need an iphone,place_order
Whats my balance,balance
Can I get a Samsung,place_order
I need to order some red pens,place_order
How much do I have,balance
How rich am I,balance
“A pizza, please”,place_order
Now that you organized them in a file, here are the steps to upload to the Intent’s files to eva:
1) First, go to the intent repository. If you want to import intents, click on the “Import” button bellow the “Create Intent” button.
2) If you want to import examples/utterances, go to the Intent you want to add the examples/utterances and click on the “Import examples/utterances” button below the “Description” field (as shown in the image below).
3) Then, you'll see a modal where you can select which file you want to import (as shown in the left side).
4) After that, if your file is valid, click on “Continue”. A progress bar will appear on the lower corner of the screen (as shown in the image below).
5) After your file is imported, this progress bar becomes a button so you can check details about your file.
You can store intents from the NLP you are using and build flows in eva.
When you click on "load", eva download all intents saved in the NLP you are using and store them in the repository. All the loaded versions are shown in a separate screen.
Those intents can be found in the intents repository. To use them, build a flow and, when selecting the “Intent” option, search in the repository the intent you want to use.
Remember that you can’t alter those intents. You can only edit their description and tags.
To change something, such as adding examples/utterances, you have to go back to the NLP and edit your intent there. After finishing, you have to go back to the NLP and train your flow. If you rename any intent in the NLP, eva will load it as a new intent. The one with the old name will disappear from the flow along the cells in sequence.
