Creating rules that support text analysis of social media data in 7.2.1
With the NLP Sample application, you can analyze text on social media websites such as Facebook, Twitter, and YouTube to obtain business intelligence that can improve the services that you offer to your customers. To analyze that text, you must configure a data set rule, a free text model rule, and a data flow rule.
- Create and configure an instance of a data set that allows you to connect with the Facebook API, Twitter API, or YouTube Data API.
You can also use one of the sample data sets (FBDataset, TwitterDS, or YouTubeDS) that are delivered with NLP Sample ruleset, but you must change its access details. These data sets do not work with the sample access details and can cause errors.
- Create and configure an instance of the free text model rule that performs text analysis of the records from the source data set. You can use the sample FTModels that is provided by the NLP Sample application.
- Create a data flow that references the free text model and processes the text-based content from the data set.
- On the Data Flows landing page, create and activate a data flow run to start processing text from Facebook, Twitter, or YouTube.
To set up your application infrastructure for analyzing content that is posted on social media, see the following tutorials:
Example data flow for text analysis
You have created and configured the rules that support text analytics in the Pega 7 Platform and started a data flow that includes a free text model to extract and analyze text-based content from social media. You can now access the NLP Sample portal to view all analyzed records one by one or generate text analytics reports that provide a visual overview of analyzed data.