Skip to main content

After all, what is AI (Artificial Intelligence)?

  • Updated

Index:

  1. But, after all, what is AI in Blip? 
  2. How to use AI?
  3. Published your Artificial Intelligence model?

 

Artificial intelligence (AI) is a branch of computer science that deals with building machines capable of simulating intelligent behavior. This topic has gained a lot of attention over the last 10 years due to 2 factors. The first is the increase in the amount of data available, mainly unstructured data (such as images, videos, text, etc.). The second is the evolution in the availability of cloud computing resources. These factors have driven investment in machine learning (or machine learning) technology by several companies.

When we talk about machine learning, problem solving techniques fall into two main groups: supervised and unsupervised learning. Generally speaking, supervised learning requires that we have the data as well as examples to train the model; in unsupervised learning, however, it is enough that we have the data for the model to work.

A topic of great relevance when it comes to AI is the area of ​​natural language processing (NLP). This area studies how a machine can understand the language of human beings. NLP is very important when we talk about chatbots, as the language is the foundation of this type of solution.

 

But, after all, what is AI in Blip?

Blip offers integration with artificial intelligence providers that do natural language processing (NLP) to understand the text sent by the user to the smart contact. In this case, we are talking about supervised learning algorithms. Therefore, this artificial intelligence needs to be configured and trained by platform users.

Made for those who want to make their chatbot smarter, the AI ​​in Blip helps create automated services that can reduce possible communication errors and deliver faster and more fluid conversations.

 

By using AI to create automated agents that understand how your customer speaks, you can experience these benefits:

 

  • More natural and fluid conversations between your brand and your customer.
  • Deliver information more easily without having to create lots of menus.
  • Reduction of communication errors that lead to the abandonment of the conversation.
  • Knowledge of the communication patterns of those who talk to your brand.

 

How to use AI?

To set up your AI, you need to choose an AI provider. An AI provider is a provider capable of delivering an artificial intelligence computational resource in the cloud, that is, a company that will host its artificial intelligence. Currently, Blip offers native integration with 3 different AI providers. There are performance differences between the three, which largely depends on the base you use for the model. The main difference for a business is the price that each provider charges, and the platform's proposal is exactly to allow the customer to change providers without difficulties.

 

The built-in providers are:

 

 

Once you've chosen your AI provider, you need to create some "intents" for your AI model (learn how to create intents). An intention is nothing more than a task or action that the user wants to perform, for example "issue a bill of exchange". In other words, they are the goals expressed in an input (message) from the customer. Creating an intention consists of determining some examples of sentences that express the same idea. In the case of the boleto example, other phrases that express the same idea would be: "generate a boleto", "I want a boleto" or "pay a boleto"

To complement your model, you can create "entities" that give your AI greater reach (learn how to create entities). Entities are parts of the sentence that represent an object or a value. Creating an entity consists of mapping products, services and contexts within the business.

If you are a more advanced user and want to use combinations of intents and entities to understand different contexts of your customers, the tool for you is content assist (learn how to use content assist). With it, it is possible to identify several sets of intentions and entities to give an even more assertive answer to your client.

Once your AI model is ready and configured, you must train and publish it (understand why train and publish your model). At this point, you also have an artificial intelligence testing window at your disposal that allows you to validate whether intents and entities are being identified correctly within example sentences (find out how to test your AI model).

 

Published your Artificial Intelligence model?

Now we have to configure the builder flow so that our smart contact starts to understand the users a lot more. For this, we have several options on how to use AI in the flow. But there is a very important issue: the evolution of its artificial intelligence. In order to be able to develop a good model, which foresees several options that its user can enter, we have to evolve our model over time.

Today BLIP has 2 important tools to help in the evolution of your model. The enhancement tool allows you to identify the phrases that have been sent to your Artificial Intelligence and add them as an example in your intentions (learn how to use the enhancement tool). In addition, you can use a test file or your customers' own examples to evaluate your model's performance and coverage (Analyze your model using the model analysis tool).

With these tools, you can create and configure your first artificial intelligence, apply it to your smart contact in an effective way that adheres to your business, and evolve it to take full advantage of the power of artificial intelligence. Thus, you can improve the relationship between your brand and your customers.

 

For more information, visit the discussion on the subject in our community or the videos on our channel. 😃

 

Related articles