wordcloud.Rmd

---
title: "Lesson 1 Summary"
author: "Jan Verbesselt"
date: "`r format(Sys.time(), '%d %B, %Y')`"
output: 
  html_document:
    theme: united
    toc: true
    toc_depth: 4
    number_sections: true
---

# Questionnaire

## A summary

[Google summary](https://docs.google.com/forms/d/1qVqb6g2sbrjCS3UWoqUG3X2ISW0FSiKsxE3y1_-RDKk/viewanalytics)

## What would you like to learn by following this course?

```{r, eval=TRUE, message=FALSE, echo = FALSE, results='hide'}
require(XML)
require(tm)
require(wordcloud)
require(RColorBrewer)
getwd()
u = "data/lesson1/Form Responses 1.html"
t = readHTMLTable(u)[[1]]
# out <- data.frame(as.character(t[,4]))

ap.corpus <- Corpus(DataframeSource(data.frame(as.character(t[,4]))))
# ap.corpus <- tm_map(ap.corpus, removePunctuation)
# ap.corpus <- tm_map(ap.corpus, tolower)
ap.corpus <- tm_map(ap.corpus, content_transformer(tolower))
ap.corpus <- tm_map(ap.corpus, function(x) removeWords(x, stopwords("english")))
ap.corpus <- tm_map(ap.corpus, stripWhitespace); #inspect(docs[1])
ap.corpus <- tm_map(ap.corpus, stemDocument)

ap.tdm <- TermDocumentMatrix(ap.corpus)
ap.m <- as.matrix(ap.tdm)
ap.v <- sort(rowSums(ap.m),decreasing=TRUE)
ap.d <- data.frame(word = names(ap.v),freq=ap.v)
table(ap.d$freq)
pal2 <- brewer.pal(8,"Dark2")
# png(fn, width=1480, height=900)
wordcloud(ap.d$word,ap.d$freq, scale=c(7,.4),min.freq=1,
max.words=Inf, random.order=FALSE, rot.per=.15, colors=pal2)
# dev.off()  
```

## What is your motivation to follow this course?

```{r, eval=TRUE, message=FALSE, echo = FALSE, results='hide'}
require(XML)
require(tm)
require(wordcloud)
require(RColorBrewer)
getwd()
u = "data/lesson1/Form Responses 1.html"
t = readHTMLTable(u)[[1]]

ap.corpus <- Corpus(DataframeSource(data.frame(as.character(t[,5]))))
ap.corpus <- tm_map(ap.corpus, removePunctuation)
# ap.corpus <- tm_map(ap.corpus, tolower)
ap.corpus <- tm_map(ap.corpus, content_transformer(tolower))
#ap.corpus <- tm_map(ap.corpus, function(x) removeWords(x, stopwords("english")))
ap.corpus <- tm_map(ap.corpus, stripWhitespace); #inspect(docs[1])
ap.corpus <- tm_map(ap.corpus, stemDocument)

ap.tdm <- TermDocumentMatrix(ap.corpus)
ap.m <- as.matrix(ap.tdm)
ap.v <- sort(rowSums(ap.m),decreasing=TRUE)
ap.d <- data.frame(word = names(ap.v),freq=ap.v)
table(ap.d$freq)
pal2 <- brewer.pal(8,"Dark2")
# png(fn, width=1480, height=900)
wordcloud(ap.d$word,ap.d$freq, scale=c(4,.4),min.freq=1,
max.words=Inf, random.order=FALSE, rot.per=.15, colors=pal2)
# dev.off()  
```

## What is your scripting experience?

```{r, eval=TRUE, message=FALSE, echo = FALSE, results='hide'}
require(XML)
require(tm)
require(wordcloud)
require(RColorBrewer)
getwd()
u = "data/lesson1/Form Responses 1.html"
t = readHTMLTable(u)[[1]]
df <- data.frame(as.character(t$G[c(-1,-2)]))
ap.corpus <- Corpus(DataframeSource(df))
# ap.corpus <- tm_map(ap.corpus, removePunctuation)
# ap.corpus <- tm_map(ap.corpus, tolower)
ap.corpus <- tm_map(ap.corpus, content_transformer(tolower))
#ap.corpus <- tm_map(ap.corpus, function(x) removeWords(x, stopwords("english")))
ap.corpus <- tm_map(ap.corpus, stripWhitespace); #inspect(docs[1])
ap.corpus <- tm_map(ap.corpus, stemDocument)

ap.tdm <- TermDocumentMatrix(ap.corpus)
ap.m <- as.matrix(ap.tdm)
ap.v <- sort(rowSums(ap.m),decreasing=TRUE)
ap.d <- data.frame(word = names(ap.v),freq=ap.v)
table(ap.d$freq)
pal2 <- brewer.pal(8,"Dark2")
# png(fn, width=1480, height=900)
wordcloud(ap.d$word,ap.d$freq, scale=c(6,.4), min.freq=1,
max.words=Inf, random.order=FALSE, rot.per=.15, colors=pal2)
# dev.off()  
```

# Excercise

## Nice solutions

```{r, eval=FALSE}
# Name: Team Lopez. Members: Peter Hooiveld & Bart Driessen
# Date: January 5, 2015

rm(list=ls())

# Import packages
# install.packages("raster")
# install.packages("maps")
library(raster)
library(maps)

north.arrow = function(x, y, h) {
  polygon(c(x, x, x + h/2), c(y - h, y, y - (1 + sqrt(3)/2) * h),
          col = "black", border = NA)
  polygon(c(x, x + h/2, x, x - h/2), c(y - h, y - (1 + sqrt(3)/2) *
                                         h, y, y - (1 + sqrt(3)/2) * h))
  text(x, y, "N", adj = c(0.5, 0), cex = 0.9)
}

# Define the function
MapMaker <- function(country, levels, province="No province specified"){
  datdir <- 'data'
  dir.create(datdir, showWarnings = FALSE)
  if(province=="No province specified"){
    adm <- getData("GADM", country = country,
                 level = levels, path = datdir)
  }
  else{
    adm <- getData("GADM", country = country,
                 level = levels, path = datdir)
    adm <- adm[adm$NAME_1 == province,]
  }
  plot(adm, bg = "dodgerblue", axes=T)
  plot(adm, lwd = 10, border = "skyblue", add=T)
  plot(adm, col = "green4", add = T)
  grid()
  box()
  name <- paste("NAME_",levels,sep="")
  invisible(text(getSpPPolygonsLabptSlots(adm),
                 labels = as.character(adm[[name]]), cex = 0.5, col = "white", font = 1))
  if(province=="No province specified"){
    mtext(side = 3, line = 1, paste("Province map of",country,sep=" "), cex = 1.1)
  }
  else{
    mtext(side = 3, line = 1, paste("Municipality map of",province,"of",country,sep=" "), cex = 1.1)
  }
  mtext(side = 1, "Longitude", line = 2.5, cex=1.1)
  mtext(side = 2, "Latitude", line = 2.5, cex=1.1)
  legend("topleft", "\n
Projection: Geographic\n
Coordinate System: WGS 1984\n
Data Source: GADM.org",adj = c(0, 0), cex = 0.6, col = "grey20",bg="transparent",box.col="transparent")
### Using legend is easier than using text in this case, due to the fact that you can stick it in a certian corner of the map (without having to supply any x or y coordinates)
  map.scale(grconvertX(0.02,"npc"), grconvertY(0.1, "npc"),metric = T, ratio=F, relwidth=0.2,cex=0.8)
  north.arrow(grconvertX(0.95,"npc"), grconvertY(0.9, "npc"),0.05)
### Another way to do this is by using the min and max coordinates from the bbox, as in: as.data.frame(bbox(adm))$min[1]*1.05, as.data.frame(bbox(adm))$min[2]*0.95; this can however lead to problems in maps that have completely different scalings; hence our use of grconvertX and grconvertY
}

# An example based on the funtion
### Two examples for a province
MapMaker("Netherlands",2,"Limburg")
MapMaker("France",2,"Champagne-Ardenne")

### Two examples for a country
MapMaker("Samoa",1)
MapMaker("Nepal",1)
#The Netherlands is possible but it takes a while, so only try that one if you really want to and have patience.
MapMaker("Netherlands",1
```

```{r, eval=FALSE}
# Name: David Scholte-Albers & Astrid Bos
# Date: 05 January 2015
rm(list = ls())
# Import packages
library(raster)

# Define the function
drawMap  <- function(countryname, level){
        datdir <- 'data'
        if (!file.exists(datdir)) { dir.create(datdir)}
        
        adm <- raster::getData("GADM", country = countryname, level = level, path = datdir)
        mapExtent <- extent(adm)
                
        #plot main parts of map
        plot(adm)
        plot(adm, bg = "skyblue4", axes=T)
        plot(adm, lwd = 3, border = "ghostwhite", add=T)
        plot(adm, col = "forestgreen", add = T)
        grid()
        box()
        
        #print labels on the map (labels of level chosen by user)
        label <- eval(parse(text=paste("adm$NAME_", level, sep="")))
        text(getSpPPolygonsLabptSlots(adm), labels = label, cex = 0.7, col = "white", font = 2)
        
        #print titles on axes
        mtext(side = 3, line = 1, paste("Map of ", countryname, sep=""), cex = 2)
        mtext(side = 1, "Longitude", line = 2.5, cex=1.1)
        mtext(side = 2, "Latitude", line = 2.5, cex=1.1)

        #add legend
        par(lheight=.5)
        text(mapExtent@xmin, mapExtent@ymin,"Projection: Geographic Coordinate System: WGS 1984 - Data Source: GADM.org", adj = c(0, 0), cex = 0.7, col = "white")
}

# An example based on that function

drawMap("Iceland", 1) #drawMap function uses country name and level as input arguments
```

```{r, eval=FALSE}
# Name: Nikoula, Siti Latifah and Tziolas Nikolaos
# Date: 6 January 2015

# Import packages
library(raster)

# Define the function

createMap <- function(country, level){
  # This function generates and plots a map automatically using country and level as inputs
  # level indicates map subdivision, nowadays the range available is between 0 and 5 as described in www.gadm.org/maps;however it can be varied among countries
  ## references <<
  ## "Global Administrative Areas." N.p., n.d. Web. 5 Jan. 2015. <http://www.gadm.org/maps>.
  #check the values of input "level"
    if ((level)!=0 & (level)!=1 & (level)!=2 & (level)!=3 & (level)!=4 & (level)!=5){
      stop("'level' has to be between '0' to '5'")
    }
  
  #specify the direction of data
  datdir <- 'data'
  dir.create(datdir, showWarnings = FALSE)
  
  #plot the map
  map <- raster::getData("GADM", country = country,level = level, path = datdir)
  plot(map,bg = "dodgerblue", axes=T)
  plot(map, lwd = 15, border = "skyblue", add=T)
  plot(map, col = "green4", add = T)
  grid()
  box()

  # Name of map title and axes
  mtext(side = 3, line = 1, paste("Map of", country), cex = 2)
  mtext(side = 1, "Longitude", line = 2.5, cex=1.1)
  mtext(side = 2, "Latitude", line = 2.5, cex=1.1)
  
  #name of provinces
  labels<-paste("map$NAME_", level, sep="")
  text(getSpPPolygonsLabptSlots(map),
  labels = eval(parse(text=labels)), cex =0.5, col = "white", font = 2)
  
  # create the legend
  par(lheight=.4)  
  legendText<-c("Projection: Geographic\n
Coordinate System: WGS 1984\n
Data Source: GADM.org")
  
  legend("topleft", # position
         legend = legendText, 
         cex = 0.7,
         col="black",
         bty = "n") # border
}

# Examples based on that function
createMap("Greece",1) # This example plots the provincial map of Greece
createMap("Greece",9) # #To test the function if the level input is larger than 5
```

## Comments

No need to:

- set the working directory
- to install R packages just load them

Extra:

- do not forget Name, Date
- filename - add your teamname_date_lastnames_   e.g. Lesson1.R is not good.

# Self reflection

* How was the excercise? How was TryR?