Examples with R

#------------------ EXAMPLE 1 --------------------------------------------------

# count the number of works in CELLAR per year

# @author : Duy Dinh

# @date: 01/06/2015

# input file containing data

input="C:/Users/dinhduy/workspace/semantic-web-tools/input/statistics/works_per_year.dat"

# read data into a table

data = read.table(input, header=TRUE)

par(mfrow=c(1,1))

maxY = max(data$TOTAL + 100)

minY = 1000

lineColors=c('blue')

plot(data$year, data$TOTAL, type = "o",

xlab="Year", ylab="Total number of works",

main="Total number of works per year in CELLAR",

col=lineColors,

ylim=c(minY, maxY)

)

#------------------ EXAMPLE 2 --------------------------------------------------

# count the number of works in CELLAR per month for a specific year

# @author : Duy Dinh

# @date: 01/06/2015

# input file containing data

input="C:/Users/dinhduy/workspace/semantic-web-tools/input/statistics/last5years"

fileFilter="*.dat"

# full file names

files = list.files(input, full.names=TRUE, pattern=fileFilter)

# short file names

names = list.files(input, pattern=fileFilter)

# number of files

n=length(files)

# distribute plots

par(mfrow=c(2,3)) # div by 2 x 3 plot

lineColors=c('blue')

pchars=c(22) # a rectangle symbol

marginY=1000

months=c('Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec')

# get the month having a particular works

# @x: a particular number of works

# @array: the set of numbers of works during a year (12 months)

getMonth <- function (x, array){

len = length(array)

for (i in 1 : len){

if (x == array[i]){

return (i) # month i found

}

}

return (array[1]) # the first month

}

# for each file, read the data into a table and plot data

for (i in 1 : n){

# read data into a table

data = read.table(files[i], header=TRUE)

maxY = max(data$TOTAL)

minY = min(data$TOTAL)

total = sum(data$TOTAL)

plot(data$month, data$TOTAL, type = "o",

xlab="Month", ylab="Nr. of works",

main=strsplit(names[i], ".", fixed=TRUE)[[1]][1],

col=lineColors,

pch=pchars,

ylim=c(minY-marginY, maxY+marginY)

)

legend("bottomright", paste("#total:", total), bty="n")

# compute the months where we observe the min or max number of works

minX = getMonth(minY, data$TOTAL)

points(minX, minY, col='red', type="o", cex=5)

maxX = getMonth(maxY, data$TOTAL)

points(maxX, maxY, col='green', type="o", cex=5)

}

#------------------ EXAMPLE 3 : SPARQL + curl over HTTP --------------------------------------------------

# This script is used to run a SPARQL query, extract the results in CSV format and the visualize data in a plot

# @author: Duy Dinh

# @date: 05/06/2015

# import libraries

library(RCurl)

# end import libraries

# year of interest

year = 2010

# SPARQL endpoint (CELLAR PROD)

endpoint <- "http://publications.europa.eu/webapi/rdf/sparql"

# ----- plot setting ----------------

marginY=100

pChars=c(22)

pColors=c('blue')

# ----- end plot setting ---------------

# User defined functions

# define the SPARQL function that sends and gets the results returned by a SPARQL endpoint

# @endpoint a SPARQL endpoint

# @query SPARQL query

# @parameters HTTP parameters associated to the query

SPARQL <- function(endpoint, query, parameters){

encodedQuery = URLencode(query, reserved = TRUE)

url=paste(endpoint, "?query=", encodedQuery, "&", params, sep="")

# encode the URL

url=URLencode(url, reserved = TRUE)

content = getURL(url,

httpheader=c('Accept'="text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8",

'User-Agent'="Google Chrome Browser"))

return (content)

}

# --------------------- MAIN PROCEDURE -------------------------------

# SPARQL query

query <- sprintf("

DEFINE input:inference \"cdm_rule_set\"

PREFIX cdm: <http://publications.europa.eu/ontology/cdm#>

PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>

PREFIX owl: <http://www.w3.org/2002/07/owl#>

PREFIX at: <http://publications.europa.eu/ontology/authority/>

SELECT ?month as ?MONTH COUNT(distinct ?work) as ?TOTAL

WHERE {

?work cdm:date_document ?date_doc .

filter ( year(?date_doc) = %d )

}

GROUP BY (month(?date_doc) AS ?month)

ORDER BY ?month

", year)

# Query parameters: format of the results returned by SPARQL endpoint

params="format=text/csv"

content = SPARQL(endpoint, query, params)

# extract data from CSV results

data = read.table(textConnection(content), header=TRUE, sep=",")

# find the maxi and mini number of works during this year

# get the month having a particular works

# @x: a particular number of works

# @array: the set of numbers of works during a year (12 months)

getMonth <- function (x, array){

len = length(array)

for (i in 1 : len){

if (x == array[i]){

return (i) # month i found

}

}

return (array[1]) # the first month

}

# global min, max number of works

maxY = max(data$TOTAL)

minY = min(data$TOTAL)

# and the total number of works during this year

total = sum(data$TOTAL)

# plot data

plot(data, type = "o", main=sprintf("Number of works per month in CELLAR during %d", year),

ylim=c(minY - marginY, maxY + marginY), pch=pChars, cex=1.5, lwd=2, col=pColors)

minX = getMonth(minY, data$TOTAL)

points(minX, minY, col='red', type="o", cex=5)

maxX = getMonth(maxY, data$TOTAL)

points(maxX, maxY, col='green', type="o", cex=5)

legend("bottomright", paste("#total:", total), bty="n")

#------------------ EXAMPLE 4: while loop --------------------------------------------------

# This script is used to run a SPARQL query, extract the results in CSV format and the visualize data in a plot

# @author: Duy Dinh

# @date: 05/06/2015

# import libraries

library(RCurl)

# end import libraries

# year of interest

year = 2011

# SPARQL endpoint (CELLAR PROD)

endpoint <- "http://publications.europa.eu/webapi/rdf/sparql"

# ----- plot setting ----------------

marginY=100

pChars=c(22)

pColors=c('blue')

# ----- end plot setting ---------------

# User defined functions

# define the SPARQL function that sends and gets the results returned by a SPARQL endpoint

# @endpoint a SPARQL endpoint

# @query SPARQL query

# @parameters HTTP parameters associated to the query

SPARQL <- function(endpoint, query, parameters){

encodedQuery = URLencode(query, reserved = TRUE)

url=paste(endpoint, "?query=", encodedQuery, "&", params, sep="")

# encode the URL

url=URLencode(url, reserved = TRUE)

content = getURL(url,

httpheader=c('Accept'="text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8",

'User-Agent'="Google Chrome Browser"))

return (content)

}

# --------------------- MAIN PROCEDURE -------------------------------

curDate = as.Date(Sys.time())

curYear = as.numeric(format(curDate, "%Y"))

year = curYear - 5

par(mfrow=c(2,3)) # div by 2 x 3 plot

while (year <= curYear) {

# SPARQL query

query <- sprintf("

DEFINE input:inference \"cdm_rule_set\"

PREFIX cdm: <http://publications.europa.eu/ontology/cdm#>

PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>

PREFIX owl: <http://www.w3.org/2002/07/owl#>

PREFIX at: <http://publications.europa.eu/ontology/authority/>

SELECT ?month as ?MONTH COUNT(distinct ?work) as ?TOTAL

WHERE {

?work cdm:date_document ?date_doc .

filter ( year(?date_doc) = %d )

}

GROUP BY (month(?date_doc) AS ?month)

ORDER BY ?month

", year)

# Query parameters: format of the results returned by SPARQL endpoint

params="format=text/csv"

content = SPARQL(endpoint, query, params)

# extract data from CSV results

data = read.table(textConnection(content), header=TRUE, sep=",")

# find the maxi and mini number of works during this year

# get the month having a particular works

# @x: a particular number of works

# @array: the set of numbers of works during a year (12 months)

getMonth <- function (x, array){

len = length(array)

for (i in 1 : len){

if (x == array[i]){

return (i) # month i found

}

}

return (array[1]) # the first month

}

# global min, max number of works

maxY = max(data$TOTAL)

minY = min(data$TOTAL)

# and the total number of works during this year

total = sum(data$TOTAL)

# plot data

plot(data, type = "o", main=sprintf("CELLAR statistics during %d", year),

ylim=c(minY - marginY, maxY + marginY), pch=pChars, cex=1.5, lwd=2, col=pColors)

minX = getMonth(minY, data$TOTAL)

points(minX, minY, col='red', type="o", cex=5)

maxX = getMonth(maxY, data$TOTAL)

points(maxX, maxY, col='green', type="o", cex=5)

legend("bottomright", paste("#total:", total), bty="n")

year = year + 1

}