Summary

In this chapter, we learned why metadata extraction is really important before looking at the use case for LiveRight, our fictitious bank, which had acquisitions that made a press release statement. Financial analysts wanted to quickly evaluate the events and entities concerning this press release and wanted to make market predictions. We looked at an architecture to help you accomplish this. In the architecture shown in Figure 1.1, we spoke about how you can use AWS AI services such as Amazon Textract to extract text from the sample press release documents. Then, we saved all the text with utf-8 encoding in the Amazon S3 bucket for Amazon Comprehend entity or metadata extractions jobs.

We used an Amazon Comprehend Events job to extract entities and relationships between the entity. We have provided a walkthrough video link of the Comprehend Events feature in the Further reading section if you wish to learn more. We also provided two ways to configure Comprehend Events job...