NCATComp410 / comp410_spring_2024

COMP410 spring 2024 semester
MIT License
3 stars 0 forks source link

Anonymized Quality #70

Open msmayaswan opened 7 months ago

msmayaswan commented 7 months ago

Discussion to explore ways to improve recognition quality.

msmayaswan commented 7 months ago

Case 1: was anonymized incorrectly for the text "52824753556AF".

Shunwel commented 7 months ago

No issues were found with IT_IDENTITY_CARD, it was correctly anonymized.

obiudemadu commented 7 months ago

US_BANK_NUMBERS was anonymized correctly, there were no issues found when running the program.

danielvelf commented 7 months ago

ES_NIF did not anonymize correctly because of errors I made during the sprint 2. After being adjusted ES_NIF was correctly anonymized and works fine now.

thirdI3 commented 7 months ago

US_LOCATION was anonymized correctly, and therefore produced no errors.

claesmk commented 7 months ago

@obiudemadu do you think the first line was a reasonable replacement for US_BANK_NUMBER ?

Customer first provided an invalid AU ABN <US_BANK_NUMBER> and then the correct one <AU_ABN>
obiudemadu commented 7 months ago

@claesmk The first line seems like a reasonable replacement for "US_BANK_NUMBER" if the context indicates that the original input was indeed an invalid Australian Business Number (ABN) rather than a US bank number.