Medical mining

This tool allows you to create and control topic models. It works best with LDA and lets you visualize the results using LDAvis.

First setup:

Checkout the project
install all packages required
start the shiny application by run app

Import your data

To import your own data, all it needs is to be a list of documents.

Minimal example:

data <- list("this is my example text","here i talk about the weather")

Edit this in the file tmscriptFacade.R

Installation

on linux (ubuntu 14.04) apt-get install libxml2-dev libmpfr4 libmpfr-dev
install gnu mp
install gnu scientific library gsl
devtools::install_github("kshirley/LDAtools")
devtools::install_github("cpsievert/LDAvis")
devtools::install_github("benmarwick/LDAviz")
install.packages("shiny")
install.packages("shinydashboard")
install.packages("topicmodels")
install.packages("servr")
install.packages("tm")
install.packages("Rmpfr")
install.packages("SnowballC")
install.packages("RJSONIO")
install.packages("RCurl")
install.packages("XML")
install.packages("stringr")
install.packages("rmongodb") # (only if you want to use MongoDB)

Note: on a mac you have to install gsl (brew install gsl)

You can also try to use the started streamgraph visualization to visualize topics along time, but this is under development.

devtools::install_github("hrbrmstr/streamgraph")