The concept of data repair comes from the realization that legacy information stores were
often created without fully verifying the input. Sometimes the stores were created by combining
multiple sources with different formats with a less than ideal result. While it is not
always possible to replace missing information, it is often possible to fix that which is
blatantly wrong. An example would be postal codes mismatched with locales. Also, it is sometimes
possible to correct data by matching it against other data sources known to be accurate.
Rather than just moving the data through simple filters, we create interface programs
with intuitive logic to scrub the data replacing known errors, bad abbreviations, or misspellings
with intelligent choices.
Call or e-mail us to discuss your needs.