How-tos

Arria brings Natural Language Generation to IBM Cloud

Share this post:

The Arria Natural Language Generation APIs service is an addition to the Finance category on the IBM Cloud platform. This blog post shows you how to get started with Arria’s Natural Language Generation APIs service on the IBM Cloud platform.

Arria logoAbout Arria’s Natural Language Generation

A leader in AI technology development, Arria delivers patented natural language generation (NLG) technologies that replicate the way a human expert analyzes data to pick out the important information, and then generate natural language to describe it.

 

Arria’s NLG solutions enjoy a substantial competitive edge in the marketplace for a few important reasons. Arria’s technologies:

  • extract key insights directly from underlying data to create unique data-driven narratives every time.
  • capture industry expertise and the interpretation strategy of the expert.
  • are based on over 30 years of data science and computational linguistics expertise.

Building NLG into IBM Cloud applications

Any application that presents information to be analyzed, whether presented in the form of a visualization, a table, or another format, can be enhanced with the addition of a narrative description explaining the key insights. With Arria’s NLG APIs, you can quickly and easily integrate NLG into your application to add context-aware, instantly updated, domain-relevant narrative explanation of any data.

In the upcoming Predictive Market Stress Testing starter kit, for example, one of Arria’s NLG APIs teams up with three other IBM Cloud platform services to generate a correlated market stress test and analysis that lets investors see what different market conditions would do to an investment portfolio:

  • One service loads the end user’s investment portfolio.
  • Another service creates scenarios to run for predicting outcomes of market changes.
  • The third runs an analysis and displays a table with current values as well as stressed values (those resulting from the chosen scenario).
  • The fourth service—Arria’s Natural Language Generation APIs service—provides generation of narrative that appears as if human-written, calling out and describing the most important facts and insights within the data.

The narrative is written from the perspective of a wealth manager or research analyst and is derived directly from the data displayed.

Using the Natural Language Generation APIs service on IBM Cloud

You can use the Natural Language Generation APIs service via the IBM Cloud console. If you don’t already have an IBM Cloud platform account, create an account now.

To find Arria’s Natural Language Generation APIs service, search the Cloud catalog for “Natural Language Generation” or navigate to it as follows:

  1. From the Cloud dashboard, click Catalog (in the top right of the screen).
  2. In the navigation at the left, under Platform, click Finance.
  3. View the tile for Arria Natural Language Generation APIs.

The next steps assume that you have already created an application in the IBM Cloud platform, and now you want to add the Natural Language Generation APIs service to it. (If you don’t already have an application, you can quickly create one by grabbing a boiler plate app from the Catalog.)

To get your credentials for using the Natural Language Generation APIs service, follow the instructions here.

Creating a narrative

The Predictive Market Stress Testing Narrative API preview dialog is the place to test out the Natural Language Generation APIs service (see below). Maybe you already tried it out when you were getting your credentials? If not, follow the instructions here to try out creating a narrative. You can use the sample JSON file provided.

Connecting the service to your application

In the IBM Cloud console, once you have added your credentials in the Service page for the Natural Language Generation APIs service, you click the Create button. After that, you should see the service in your Dashboard.

Now you connect the Natural Language Generation APIs service to your application, then re-stage your application.

  1. Click the service name in your Services list.
  2. In the information page for the service, navigate to Connections and click the Create Connection button.
  3. In the row for your application, click the Connect button.
  4. In the Re-stage App dialog, click Restage to re-stage your application with the new service connection.
    When re-staging is finished, you see a tile in the information page for the Natural Language Generation APIs service. The tile is for your own application, and it means that your application is now connected to the Natural Language Generation APIs service.
  5. At the bottom of your application’s tile, click Start. Now your re-staged application is up and running again.

Next steps

Now that you have the Natural Language Generation APIs service in your application and you have tried out creating narrative output from sample data, you can decide how you want to use NLG in your application.

For more information about the features of Arria’s Natural Language Generation APIs:

For questions about NLG, or for technical questions, please visit our Support page for instructions on how to contact us.

 

More How-tos stories

How to Backup Your IBM Cloud Linux Server

This post covers how to backup entire partitions in Linux systems. The process utilizes the tar command in IBM Cloud's unique rescue environment and showcases the simplicity and flexibility of the process.

Continue reading

Speed up your WordPress with IBM Cloud

WordPress is one of the most popular content management systems available, but the many websites and blogs that use it experience issues with speed. At IBM Cloud, there are several solutions that can help alleviate some of these issues and allow you to have a better and faster WordPress experience.

Continue reading

Container Native Monitoring Insights with Elastic on IBM Cloud

Learn about how Elastic easily deploys with the IBM Cloud Kubernetes Service (IKS), providing full visibility of your containerized workloads and operational consistency with container deployments in a multi-cloud architecture.

Continue reading