chunk_separation_vs.Rmd

---
title: "R Notebook"
output: html_notebook
---


Load data
```{r}
dat<-read.csv("~/Documents/NYU/Fall 2017/Text Analysis Project/cpsv_text_project/chat_time_series.csv")

dat$obs <- seq.int(nrow(dat))

dat<-dat[,-1]
```

Seperate the chunk for all groups
```{r}
#create new dataset where I am going to add one more column called "chunk ID"
#wchunk<- matrix(NA, nrow = 21829, ncol = 9)

obsnn<- rep(NA,21829)
#dat_obsnn<- matrix(NA, nrow=21829, ncol = 2)


dat$chunk_id<- NA

#not run
for (i in 1: length(unique(dat$group_id))){
  for (j in (1: length(dat$obs))){
    dat_obsnn[i,1]<-
    dat_obsnn[j,2]<- seq.int(nrow(dat))
  }
}


for (i in (1:length(dat$obs))){
 dat_obsnn[i]<- dat$obs[which(group2$module==grp2_module[i])] 
}

for (i in 1: length(unique(dat$group_id))){
    dat$obsnn[i]<-seq.int(unique(dat$group_id)[i]==dat$group_id)
}

#not run end

temp<-list()
dat$module_id<- NA

grp<- unique(dat$group_id)
for (i in 1: length(grp)){
    temp[[i]]<-seq.int(which(dat$group_id==grp[i]))
}
temp_list<- unlist(temp)

dat$obs_grp<- temp_list

dat_obsnn<- matrix(NA, nrow=21829, ncol=9)

dat_obsnn[,1]<- dat$group_id
dat_obsnn[,2]<- dat$module
dat_obsnn[,3]<- dat$obs_grp
dat_obsnn[,4]<- dat$user_id
dat_obsnn[,5]<-dat$content
dat_obsnn[,6]<-dat$correct

temp2<- dat_obsnn[,2]

dat$obsnn<- temp2

print(dat$obsnn[which(dat$obsnn != "NA" & dat$group_id==1)])

#not run

#for (i in (1: nrow(dat))){
#  for (j in (1: length(grp))) {
#    idx[j]<- dat$obsnn[which(dat$obsnn != "NA" & dat$group_id==j)]
#      if(duplicated(idx[j] |duplicated(idx[j]),fromLast = TRUE)==TRUE){
#    dat$chunk_id[i]<- dat$obsnn[which(dat$obsnn != "NA" & dat$group_id==j)][2]
#  }
#  else {
#    dat$chunk_id[i]<- dat$obsnn[which(dat$obsnn != "NA" & dat$group_id==j)][1]
#  }
#  }
#}
#not run end
dat$chunk_id<- dat$obsnn
dat$chunk_id<- na.locf(dat$chunk_id,na,rm=TRUE, fromLast = TRUE)

View(dat)

write.csv(dat, "chunk_seperated file.csv")




```