rflynn / pro-file

statistical inferences from common online profile info

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Pro-File
-------------------------------------------------------------------------------

Statistical profiling of commonly-available personal information.

For example, a given first name may suggest gender, age and ethnicity.
These suggestions then suggest yet more things such as education level,
interests, BMI, height, etc.

We can't possibly know everything about someone from a name and an IP address,
but this sort of information _can_ give us a starting point for customization
of an online experience which can ultimately result in a happier customer.

We'll construct a database using accurate publicly-available census, genealogical 
and other data and and then cross-reference an incomplete set of information and
return a set of statistical speculations.

Requirements:
	sudo apt-get install -y sqlite3 python3

Use:

	make

	This will download data files and use them to generate a local database.
	It may take a few minutes; resulting db should be ~100MB.

	in data/
		./givenname.py "my first name"
		./surname.py "my last name"

About

statistical inferences from common online profile info


Languages

Language:Python 98.8%Language:Makefile 0.6%Language:PLpgSQL 0.4%Language:Batchfile 0.3%