https://github.com/chainsawriot/hkrug_tm

Hong Kong R User Group Pre-hackathon workshop on Text Mining.
https://github.com/chainsawriot/hkrug_tm

Last synced: 27 days ago
JSON representation

Hong Kong R User Group Pre-hackathon workshop on Text Mining.

• Host: GitHub
• URL: https://github.com/chainsawriot/hkrug_tm
• Owner: chainsawriot
• License: gpl-2.0
• Created: 2015-03-26T13:25:08.000Z (over 9 years ago)
• Default Branch: master
• Last Pushed: 2015-04-16T03:56:25.000Z (over 9 years ago)
• Last Synced: 2023-03-23T17:29:21.905Z (over 1 year ago)
• Language: R
• Size: 556 KB
• Stars: 3
• Watchers: 2
• Forks: 3
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# hkrug_tm

Hong Kong R User Group Pre-hackathon workshop on Text Mining.

This pre-hackathon is designed for those who want to get their hands dirty to build something. This is not a chit-chat session (in cantonese 吹水). YOU NEED TO CODE DURING THE WORKSHOP.

# prerequsite

You need to know some R and basic stats (regression and clustering analysis). You need to know what is going on with the following two R-snippnets.

```{r}

#R-code snippet 1

require(MASS); summary(glm(as.factor(vs)~mpg+cyl, data=mtcars, family=binomial)) ; predict(glm(as.factor(vs)~mpg+cyl, data=mtcars, family=binomial), mtcars)

#R-code snippet 2

kmeans(iris[,1:4], 3)

kmeans(iris[,1:4], 3)$cluster

table(kmeans(iris[,1:4], 3)$cluster, iris[,5])

```

# preparation

You should be running R > 3.0 and your preferred editor. (RStudio/emacs/vim/sublime) Please install the required R packages

```{r}

install.packages(c("tm", "stringr", "plyr", "topicmodels", "SnowballC", "magrittr", "klaR", "e1071"))

```

For those who are using Linux, you need to install GNU scientific library to compile the "topicmodels" library. If you are using Debian/Ubuntu, you can install GNU scientific library by:

```

sudo apt-get install libgsl0ldbl libgsl0-dev

```