Hi,
I need kind help from you. I'm doing my assignment in IR and need to do script
in R programming and using R studio tool.I don't have any knowledge in R but
learning by Youtube. After so long,i successfully came out with below script
for precision 10(P@)
But i don know how to do loop for 105 system. below script was for 5 system but
i define it one by one.Can you help me ?
#
# FIND precision@10 from a vector of files for
# one topic (vector of file names)
#
##############################
## GLOBAL VARIABLES
##############################
qrelsFilePath="D:\\UM Studies_Master_MCS\\Semester_2\\WMGA6320_Special Topics
In IS\\Group_Project\\qrels.trec9.main_web"
findPrecisionAtTen
<-function(currentTopic,systemName,fileNamesForTopic,qrelsMatrix){
# list of documents for specific topic
# Qrels file
relevantCount<-0
for (j in (1:10))
{
if(length(fileNamesForTopic) >= j)
{
# Get the corresponding entries
# for that document ,and then match it
# with topic number to check its
# relevancy
# filter according to file name
files_index <- apply(qrelsMatrix, 1, function(row) any
(fileNamesForTopic[j] == row))
# filter teh retrieved possibilities to
# topic number
# note: 4 means one entry..
if(length(qrelsMatrix[files_index,])!=4 &&
length(qrelsMatrix[files_index,])!=0){
#if you reach here means you have more than one entry
topics_index<-apply((qrelsMatrix[files_index,]),1,function(row) any
(currentTopic==row))
if(length(qrelsMatrix[files_index,][topics_index,])!=0)
{
#check if relevant or not
if(qrelsMatrix[files_index,][topics_index,4]==1){
# it is relevant document
relevantCount=relevantCount+1
}else{
#not relevant ->do nothing
}
}else{
#do nothing it is not relevant
}
}else{
# one entry or none..
if(length(qrelsMatrix[files_index,])==4) #one option
{
#only one entry found -> check the topic and relevancy together.
if(qrelsMatrix[files_index,1]==currentTopic &&
qrelsMatrix[files_index,4]==1){
#relevant
relevantCount=relevantCount+1
}else{
# not relevant
}
}else{
#not in the file -> none relevant
}
}
}else{
#that means that system retrieved less than 10 documents for the topic!
# not relevant :)
}
}
#print("current System :")
#print(systemName)
#print("current Topic : ")
#print(currentTopic)
#print("precision@10 : ")
return (relevantCount/10)
}
#
# START FROM HERE (Main Function)
#
findPrecsionForTopicsList <-
function(qrelsFilePath,runFilePath,systemName,startTopic,endTopic){
# read Qrels File.
qrels<-read.table(qrelsFilePath, sep=" ")
qrelsMatrix<-as.matrix(qrels)
# read System run File.
run<-read.table(runFilePath, sep=" ")
runMatrix<-as.matrix(run)
currentTopic=startTopic#endTopic
# Iterate over all the required Topics
# and find the corresponding documents array (retrieved by the System)
# and use that array and qrels to find P@10 for that topic
results <- c(1)
i<-1
while(currentTopic <= endTopic){
#filter the documents according to the Topic
index <- apply(runMatrix, 1, function(row) any(currentTopic == row))
topicFiles<-(runMatrix[index,3])
answer<-findPrecisionAtTen(currentTopic,systemName,topicFiles,qrelsMatrix)
results[i]=answer
currentTopic <- currentTopic+1
i<-i+1
}
# record the mean of the topics precission@10
results[i]=mean(results)
return(results)
}
initilizeMatrix<-function(start,end,systemsIncluded){
topicStart=start
topicEnd=end
systemsIncluded=systemsIncluded
resultMatrixHieght= topicEnd - topicStart + 3
resultMatrixWidth= systemsIncluded + 1
x <- matrix(1:1, ncol = resultMatrixWidth,nrow= resultMatrixHieght)
x[1,1]="T/S"
# now fill the corresponding
# topics in right places
entry=topicStart
for(j in (2: resultMatrixHieght)){
x[j,1]=entry
entry=entry+1
}
x[resultMatrixHieght,1]="Avg"
#########################
# end initilize matrix#
#########################
return(x)
}
A <-function(start,end){
#########################
### START A #
#########################
x<-initilizeMatrix(start,end,5)
topicStart=start
topicEnd=end
perSystemPerTopicFileLocation_csv="D:\\UM
Studies_Master_MCS\\Semester_2\\WMGA6320_Special Topics In
IS\\Group_Project\\results.csv"
perSystemPerTopicFileLocation_txt="D:\\UM
Studies_Master_MCS\\Semester_2\\WMGA6320_Special Topics In
IS\\Group_Project\\results.txt"
###########################
#start system Number 1#
##########################
systemNumber=1
systemName="input.acsys9mw0"
systemFilePath="D:\\UM Studies_Master_MCS\\Semester_2\\WMGA6320_Special
Topics In IS\\Group_Project\\input.acsys9mw0"
result<-findPrecsionForTopicsList(qrelsFilePath,systemFilePath,systemName,topicStart,topicEnd)
x[1,systemNumber+1]=systemName
b<-1
for( j in (2:length(result))){
x[j,systemNumber+1]=result[b]
b<-b+1
}
x[j+1,systemNumber+1]=result[b]
###########################
#end system Number 1#
##########################
###########################
#start system Number 2#
##########################
systemNumber=2
systemName="input.att0010gb"
systemFilePath="D:\\UM Studies_Master_MCS\\Semester_2\\WMGA6320_Special
Topics In IS\\Group_Project\\input.att0010gb"
result<-findPrecsionForTopicsList(qrelsFilePath,systemFilePath,systemName,topicStart,topicEnd)
x[1,systemNumber+1]=systemName
b<-1
for( j in (2:length(result))){
x[j,systemNumber+1]=result[b]
b<-b+1
}
x[j+1,systemNumber+1]=result[b]
###########################
#end system Number 2#
##########################
###########################
#start system Number 3#
##########################
systemNumber=3
systemName="input.CWI0000"
systemFilePath="D:\\UM Studies_Master_MCS\\Semester_2\\WMGA6320_Special
Topics In IS\\Group_Project\\input.CWI0000"
result<-findPrecsionForTopicsList(qrelsFilePath,systemFilePath,systemName,topicStart,topicEnd)
x[1,systemNumber+1]=systemName
b<-1
for( j in (2:length(result))){
x[j,systemNumber+1]=result[b]
b<-b+1
}
x[j+1,systemNumber+1]=result[b]
###########################
#end system Number 3#
##########################
###########################
#start system Number 4#
##########################
systemNumber=4
systemName="input.dcu00ca"
systemFilePath="D:\\UM Studies_Master_MCS\\Semester_2\\WMGA6320_Special
Topics In IS\\Group_Project\\input.dcu00ca"
result<-findPrecsionForTopicsList(qrelsFilePath,systemFilePath,systemName,topicStart,topicEnd)
x[1,systemNumber+1]=systemName
b<-1
for( j in (2:length(result))){
x[j,systemNumber+1]=result[b]
b<-b+1
}
x[j+1,systemNumber+1]=result[b]
###########################
#end system Number 4#
##########################
###########################
#start system Number 5#
##########################
systemNumber=5
systemName="input.Flab9atd2N"
systemFilePath="D:\\UM Studies_Master_MCS\\Semester_2\\WMGA6320_Special
Topics In IS\\Group_Project\\input.Flab9atd2N"
result<-findPrecsionForTopicsList(qrelsFilePath,systemFilePath,systemName,topicStart,topicEnd)
x[1,systemNumber+1]=systemName
b<-1
for( j in (2:length(result))){
x[j,systemNumber+1]=result[b]
b<-b+1
}
x[j+1,systemNumber+1]=result[b]
###########################
#end system Number 5#
##########################
# write the results You got to files...
write.table(x, file = perSystemPerTopicFileLocation_csv,
append = FALSE, sep = ",",col.names = F, row.names = F)
write.table(x, file = perSystemPerTopicFileLocation_txt,
append = FALSE, sep = "\t",col.names = F, row.names = F)
#########################
### END A #
#########################
}
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
