Class 8
The ETL Process: Text Analytics
Description
This session introduces methods for transforming text into quantitative measures that can be analyzed systematically. Text analytics is important because a substantial portion of financially relevant information appears in disclosures, reports, correspondence, and other narrative sources rather than traditional numerical datasets.
Motivation
How do we quantify language?
Review
Regular expressions allowed us to identify known patterns and extract specific pieces of information from text. This session moves beyond locating predefined patterns and considers how the broader characteristics and content of language can themselves become data.