Case Study
Semi-Structured Text Annotation
Challenge
A seemingly simple process of parsing data from a PDF is made increasingly difficult when the format of the document changes frequently. It is time-consuming and expensive to keep the computer model updated to extract the data requested.
Industry
Technology
Data Type
Project Duration
2 Months
Ongoing?
Yes
Solution
Our Data Associates reviewed the documents and used context clues to extract the desired information from each PDF quarterly report, navigating nuanced different presentations of the information.
Outcome
The resulting data supports tracking consumer trends throughout the current year as well as YearOver-Year comparisons. This makes various analyses of the data easier, more consistent, and standardized across a variety of different documents.
Download Case Study
