Project Debater is the first AI to successfully engage with a human in a live debate. In February 2019, Project Debater debated Mr. Harish Natarajan, one of the world’s leading professional debaters in an event that was held in San Francisco and broadcast live worldwide. The underlying technologies that enabled the live event are now available to you as software services, denoted as Project Debater services. Project Debater services include:
- Natural language understanding capabilities
- Argument mining
- Text summarization
- Narrative generation
You can get more information on the various services, code examples, and online demos at the IBM Debater Early Access Program site.
When you have a large collection of texts representing peoples’ opinions (such as product reviews, survey answers, or posts from social media), it is difficult to understand the key issues that come up in the data. Going over thousands of comments is prohibitively expensive. Existing automated approaches are often limited to identifying recurring phrases or concepts and the overall sentiment toward them, but do not provide detailed or actionable insights.
In this tutorial, get hands-on experience using Project Debater services for analyzing and deriving insights from open-ended answers. The data you use is from a community survey that was conducted in 2016 and 2017 in Austin, Texas. You’ll analyze the open-ended answers in different ways by using four Debater services: the Argument Quality service, the Key Point Analysis service, the Term Wikifier service, and the Term Relater service. Then, you’ll see how they can be combined into a powerful text analysis tool.
To follow this tutorial, you first need to receive credentials for the IBM Debater Early Access Program site by sending a request email to firstname.lastname@example.org.
It should take approximately 1 hour to complete the tutorial.
Step 1. Environment setup
To set up your environment:
Create a new conda environment with the following command.
conda create --name environment_name python=3.7
Activate the new environment.
conda activate environment_name
Log in to the Debater APIs Eary Access Program site using the credentials that you received (see Prerequisites).
Download and install the SDK using the instructions at https://early-access-program.debater.res.ibm.com/download_sdk.html.
Check that the environment is set up correctly by running a code example. See https://early-access-program.debater.res.ibm.com/#keypoints for more information.
Get the API-KEY from the site.
Step 2. Tutorial setup
To install the Debater tutorial:
Get the tutorial.
git clone https://github.com/IBM/debater-eap-tutorial.git
Install the tutorial’s requirements.
cd debater-eap-tutorial pip install -r requirements.txt
Step 3. Run the tutorial
Run the Jupyter Notebook (use the api-key from the site).
env DEBATER_API_KEY="PUT_YOUR_API_KEY_HERE" jupyter notebook
Open the survey_usecase/austin_demo.ipynb notebook for a simple hands-on tutorial or open survey_usecase/austin_demo_with_answers.ipynb for a version that you can run.
This tutorial explained how to install, set up, and run the Project Debater service tutorial. In that tutorial, you got a hands-on experience using Project Debater services for analyzing and deriving insights from open-ended answers. If you have any questions or issues, please contact the Project Debater team at email@example.com.