
Chapter 12 Unstructured Business Metadata
224
12.3.1 A Distillation
Once the text is pulled from the unstructured file, the text can be examined. The
starting point for an examination of the text—a distillation of unstructured text—is
to pull recognized words out of the file. Figure 12.4 shows this extraction of words.
Once the words are pulled out of the file, they are separated into two categories—
words that are business relevant and those that are not relevant. Only words that are
business relevant are of interest to the developer building the infrastructure.
A further distinction can be made as to whether the words should be clas-
sified as business ...