Color coding of text to make looking for important (but unknown words) easier, i.e. Personally Indentifiable Information (PII) in freeform text.
Definitely not in a working state at the moment. Web dev is not my strong point so I'm using this project to learn a bit about it.
Humans are great at pattern recognition, computers not as much. So, leverage human pattern recognition to find words in text that shouldn't be there. Over time build up a list of known good and known bad strings. This can, at some future point, be used as a training set for machine learning.
The current use case is finding PII (names, SS#'s, phone #'s, addresses, etc) in freeform text. As a developer syntax highlighting is a huge help when I'm reading or writing code. There's no reason why we can't provide a syntax highlight for known good or bad words.
- Import a csv of words.
- It's a specific format currently, making it dynamic will be a future enhancement.
- Import a list of words that are known good.
- These words will be greyed out so the eye skips over them easier.
- In my particular use case, this can be provided via a text of the 1k most common english language words.
- Import a list of words that are known bad.
- These words will be red so they eye picks them up easier.
- This color should be configurable in the future for different applications/color blindness/cultural color differences.
- All other words will be "standard" black.
- Each word is clickable to cycle all instances of that word through the 3 states.
- Each row of text is clickable so it can be exported for processing in another application.
- In essence this is to mark a row as containing highlighted red words.
- Use css classes for color coding
- Database Access
- Export of Text to Summarize
- Integration with Ticketing System