Edit

Innovate with AI in Azure

  • Article

As an innovator, your company has rich information about its business and its customers. Using AI innovation, your company can:

  • Make predictions about customer needs.
  • Automate business processes.
  • Discover information that's latent in unstructured data.
  • Engage with customers in new ways to deliver better experiences.

This article introduces a few approaches to innovating with AI. Innovations can expand your company's business insights into your existing data. The following table can help you find the best solution for your implementation needs.

Solution category Description Required skills
Machine learning Azure Machine Learning
Build, deploy, and manage your own machine learning models.		 Data scientist and developer
AI applications and agents Azure Cognitive Services
Use domain-specific AI models for vision, speech, language, and decision that can be customized with your data.

Azure Bot Service
Improve customer engagement by adding bots to your applications and websites.		 Developer
Knowledge mining Azure AI Search
Uncover insights that are latent in your content, including documents, contracts, images, and other data types.		 Developer

Machine learning

Azure provides advanced machine learning capabilities. Build, train, and deploy your machine learning models across the cloud and edge by using Azure Machine Learning. Develop models faster by using automated machine learning. Use tools and frameworks of your choice without being locked in.

For more information, see Azure Machine Learning overview and Getting started with your first machine learning experiment. For more information on the open-source model format and runtime for machine learning, see ONNX Runtime.

Action

A data scientist can use Azure Machine Learning to train and build a model by using advanced languages such as Python and R, as well as by using a drag-and-drop visual experience. To get started with Azure Machine Learning:

  1. In the Azure portal, search for and select Machine Learning.

  2. Select Add, and follow the steps in the portal to create a workspace.

  3. The new workspace provides both low-code and code-driven approaches for data scientists to train, build, deploy, and manage models.

Go directly to Azure Machine Learning resources in the Azure portal.

AI applications and agents

Azure provides a set of prebuilt AI services called Cognitive Services to build AI applications. Additionally, Azure offers bot service, which allows developers to build conversational AI agents that improve customer and employee engagement.

AI applications

Cognitive Services enables you to incorporate the AI capabilities of vision, speech, language, and decision into your applications. Most predictive models don't require additional training. These services are useful when you don't have data scientists on staff to train the predictive model. Other services require minimal training.

For more information about the training that might be required and a list of available services across vision, speech, language, and decision-making, see the Cognitive Services documentation.

Action

To get started with a Cognitive Services API:

  1. In the Azure portal, search for and select Cognitive Services.

  2. Select Add to find a Cognitive Services API in Azure Marketplace.

  3. Search for and select a service:

    • If you know the name of the service you want to use, enter the name in Search the Marketplace. Then select the service.

    • For a list of Cognitive Services APIs, select See More next to the Cognitive Services heading. Then select the service.

  4. Select Create, and follow the steps in the portal to provision the service.

Go directly to Cognitive Services in the Azure portal.

AI agents

Interact more naturally with your customers and improve customer engagement through conversational experiences powered by Bot Framework and Azure Bot Service. In addition, use Cognitive Services APIs like Language Understanding (LUIS), QnA Maker, and Speech service. These help your customers with common tasks, leaving your call center agents time to focus on more nuanced, higher value cases.

For more information on how to build bots, see Azure Bot Service.

Action

To get started with Azure Bot Service:

  1. In the Azure portal, search for and select Bot Services.

  2. Select Add, and then select web app bot or bot channels registration.

  3. Select Create. Then follow the steps in the portal to provision the service.

Go directly to Azure Bot Service in the Azure portal.

Knowledge mining

Knowledge mining uses AI to drive content understanding over vast amounts of unstructured, semi-structured, and structured information. Use Azure AI Search to uncover latent insights from your content, including documents, images, and media. You can discover patterns and relationships in your content, understand sentiment, and extract key phrases.

Azure AI Search uses the same natural language stack that Bing and Microsoft Office use. Spend more time innovating and less time maintaining a complex cloud search solution.

For more information, see What is Azure AI Search?

Action

To get started:

  1. In the Azure portal, search for and select Azure AI Search.

  2. Follow the steps in the portal to provision the service.

Go directly to Azure AI Search in the Azure portal.