andrie

## Analyze R packages for popularity, using pagerank algorithm

# Inspired by Antonio Piccolboni, http://piccolboni.info/2012/05/essential-r-packages.html

library(miniCRAN)
library(igraph)
library(magrittr)


# Download matrix of available packages at specific date ------------------

andrie

# Computing the probability of drawing two identical
# numbers out of a hat



# Incorrect - binomial distribution ---------------------------------------

# Use pbinom() to compute a probability curve for binomial 
# distribution

andrie

## Determine CRAN package clusters (communities)


library(miniCRAN)
library(igraph)
library(magrittr)


# Download matrix of available packages at specific date ------------------

andrie

# Wrapper around available.packages ---------------------------------------

index <- function(url, type="source", filters=NULL){
  contribUrl <- contrib.url(url, type=type)
  available.packages(contribUrl, type=type, filters=filters)
}


# CRAN --------------------------------------------------------------------

andrie

library(foreach)
library(iterators)
library(doParallel)
library(tcltk)

# Choose number of iterations
n <- 250

# In sequence, without progress bar ---------------------------------------

andrie

library(foreach)
library(iterators)
library(doParallel)
library(tcltk)

# Choose number of iterations
n <- 1000

cl <- makeCluster(8)

andrie

# Based on archimedes spiral
# Inspired by https://www.youtube.com/watch?v=SEiSloE1r-A
# The surprising beauty of mathematics | Jonathan Matte | TEDxGreensFarmsAcademy
# See http://en.wikipedia.org/wiki/Archimedean_spiral

rotate <- function(theta){
  matrix(
    c(cos(theta), -sin(theta), 
      sin(theta), cos(theta)
    ), 

andrie

library("networkD3")
library("igraph")

# Download prepared igraph file from github
url <- "https://github.com/andrie/cran-network-structure/blob/master/pdb/depGraph-CRAN.rds?raw=true"
datafile <- tempfile(fileext = ".rds")
download.file(url, destfile = datafile, mode = "wb")
gs <- readRDS(datafile)

# Remove all nodes with fewer than 50 edges

andrie

# StackOverflow query at 
# https://data.stackexchange.com/stackoverflow/query/338993/r-trends-questions-per-tag-per-month#resultSets


# Import data -------------------------------------------------------------

url <- "https://data.stackexchange.com/stackoverflow/csv/437426"
dat <- read.csv(url)
dat$Month <- as.Date(dat$Month)
head(dat)

andrie

# StackOverflow query at 
# https://data.stackexchange.com/stackoverflow/query/338993/r-trends-questions-per-tag-per-month#resultSets


# Import data -------------------------------------------------------------

url <- "https://data.stackexchange.com/stackoverflow/csv/440417"
dat <- read.csv(url)
dat$Month <- as.Date(dat$Month)
head(dat)