Simple Cherrypy based dataset browser for China Study II data-set
This is a work in progress. A simple cherrypy server serving up poly fit regression lines for any two variables from the China Study II dataset.
The file CS89.JSON contains the original data. A simple script extra.py processes this file and adds a few composite variables such as the probability to live to the age of at least 80 years old, or the omega6 / omega 3 ratio. The resulting data is combines in NW89.JSON, the data file used by the server script.
The script will run a basic webserver on port 8080, where it will allow to pick any two variables from the dataset together with the degrees of freedom to use in plotting a polifit regression line.
It's a real simple straightforward script at the moment and it does not do any analysis, it is just meant as a visualisation tool for the CS-II data.
I've published the code of this server in an act of defense, showing I'm not doing anything strange with the data that makes the equally named book look bad. It is all just visualisation of data that is there, no fancy stats tricks designed to make the veganism or proteinophobia look badly, just straight forward code, showing that without the weapons of eloquence and multi-layered inductive reasoning, two attributes that the books author abundantly posesses, the actual data shows an often completely opposite picture from the one described in the book.
So to you vegan militants accusing me of misrepresenting the data, see for yourself, it really is only visualisation, this is actually how the data looks, sorry about that.
If you want to check out the running code, I have it running on a C.H.I.P (might not be up 24/7 though).
I would very much like to extent this server into a more generic epi data univar browser, but unfortunately I'm running into (re-)sharing policies for data sets. For this project to work with different data sets, I need to be allowed to convert the original data-set to a specific unified JSON format used in this project, and I need to be allowed to re-share my JSON representation of the data from this github repo. If you happen to know of any nutritional epi data set that fits these requirements that might be a usefull addition to this project, please drop me a message at pibara@gmail.com