Now the processing parameters for Textual ETL are created. These parameters may include such things as:
· External categorizations
· Stop word processing
· Alternate spelling
· Homograph resolution
· Proximity analysis
· Pattern searching, and so forth.
Once the parameters are specified, the execution panel is prepared, the queue for processing is established, and processing commences. It is absolutely abnormal for the first setting of parameters to be correct. There are so many nuances to text and so many possibilities, that an accurate initial parametric setting is almost unheard of. Instead parameters are set, a few documents are run, the results ...