I'm dealing with data that descripe onset frequency of a symptom. The text in each cell was not in the same format. For example:
"Pattern 1: 1-3/months, Pattern 2: 6 times per month"
"4-5 times/day"
"10 times/day to 7-8 times/week"
"varies (average 0-10 fits/month)"
"1-2 times/month, but 2-3/week in recent 3 months\n"
"2-3/month"
What I'm trying to do is to generate the minimum and maximum frequency of each description for further statistical analysis. I can try to complete this job by regular expression, although it might not be very efficient. I'm very new to natural lauguage processing, is there a more efficient way to do this other than the regular expression way? Thanks!
I can read and type the frequency by hand. But from what I've learned, this is always a bad practice to data analysis.
(Sorry about not being able to ask the question in a more specific way.)