Related
I would like to aggregate, in order to reduce the number of constructs, its following data frame containing only binary variables that correspond to "yes/no", its following data frame (first 10 row). The original data frame contains 169 rows.
outcome <-
structure(list(Q9_Automazione.processi = c(0, 0, 0, 0, 0, 0,
1, 1, 1, 0), Q9_Velocita.Prod = c(1, 0, 0, 1, 0, 0, 1, 1, 1,
0), Q9_Flessibilita.Prod = c(0, 0, 0, 1, 0, 0, 1, 1, 0, 1), Q9_Controllo.processi = c(0,
0, 0, 1, 0, 0, 1, 1, 0, 0), Q9_Effic.Magazzino = c(0, 0, 0, 1,
0, 0, 0, 0, 0, 0), Q9_Riduz.Costi = c(0, 1, 0, 0, 0, 0, 0, 0,
0, 1), Q9_Miglior.Sicurezza = c(0, 0, 0, 0, 0, 0, 1, 0, 1, 1),
Q9_Connett.Interna = c(0, 0, 0, 0, 0, 0, 0, 1, 1, 0), Q9_Connett.Esterna = c(0,
0, 0, 0, 0, 0, 0, 0, 0, 0), Q9_Virtualizzazione = c(0, 0,
0, 0, 0, 0, 0, 0, 0, 0), Q9_Innov.Prod = c(0, 0, 0, 0, 0,
1, 0, 0, 0, 1), Q9_Person.Prod = c(0, 1, 0, 1, 0, 1, 0, 0,
0, 1), Q9_Nuovi.Mercati = c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0),
Q9_Nuovi.BM = c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0), Q9_Perform.Energ = c(0,
0, 0, 0, 0, 0, 0, 0, 0, 0), Q9_Perform.SostAmb = c(0, 0,
0, 0, 0, 0, 0, 0, 0, 0)), row.names = c(NA, 10L), class = "data.frame")
I have tried performing factor analysis via the tethracoric method on the obtained correlation matrix ( the obtained value from the KMO function turns out to be inadequate) both directly on the dataframe and then using tethracoric correletions in fafunction (using cor = "tet" I get a negative Tucker Lewis Index).
I have been reading up on this but cannot find a methodology that is adequate and of which I am certain of the correctness of the analysis.
So basically what I would like to achieve is to aggregate similar constructs, e.g., assess whether column 5 has value 1 (i.e., "yes") almost always when column 11 has value 1 and then aggregate.
Here the code that I try to used
library(psych)
tet <- tetrachoric(outcome)
corrplot(tet$rho, "ellipse", tl.cex = 0.75, tl.col = "black")
par(mfrow = c(1,2))
corr_matrix %>%
ggcorrplot(show.diag = F,
type="lower",
lab=TRUE,
lab_size=2)
KMO(corr_matrix)
cortest.bartlett(corr_matrix)
fa.parallel(corr_matrix, fm = "ml")
factor <- fa(corr_matrix, nfactors = 3, rotate = "oblimin", fm = "ml")
print(factor, cut = 0.3, digits = 3)
# -------- Pearson --------
cor(outcome, method = 'pearson', use = "pairwise.complete.obs") %>%
ggcorrplot(show.diag = F,
type="lower",
lab=TRUE,
lab_size=2)
KMO(outcome)
cortest.bartlett(outcome)
fa.parallel(outcome)
factor1 <- fa(outcome, nfactors = 3, rotate = "oblimin", cor = "tet", fm = "ml")
print(factor1, cut = 0.3, digits = 3)
I'm trying to create a bar plot using ggplot2 and my data is in this format:
dput here:
structure(list(clade = structure(c(1L, 3L, 2L, 3L, 2L, 2L), .Label = c("19A",
"20A", "20B", "20E (EU1)", "20I (Alpha, V1)", "20J (Gamma, V3)",
"21J (Delta)"), class = "factor"), C.T = c(0, 4, 4, 4, 4, 4),
A.G = c(0, 1, 1, 1, 1, 1), G.A = c(0, 2, 0, 2, 0, 0), G.C = c(0,
1, 0, 1, 0, 0), T.C = c(0, 0, 0, 0, 0, 0), C.A = c(0, 0,
0, 0, 0, 0), G.T = c(0, 0, 0, 0, 0, 0), A.T = c(0, 0, 0,
0, 0, 0), T.A = c(0, 0, 0, 0, 0, 0), T.G = c(0, 0, 0, 0,
0, 0), A.C = c(0, 0, 0, 0, 0, 0), C.G = c(0, 0, 0, 0, 0,
0), A.del = c(0, 0, 0, 0, 0, 0), TAT.del = c(0, 0, 0, 0,
0, 0), TCTGGTTTT.del = c(0, 0, 0, 0, 0, 0), TACATG.del = c(0,
0, 0, 0, 0, 0), AGTTCA.del = c(0, 0, 0, 0, 0, 0), GATTTC.del = c(0,
0, 0, 0, 0, 0)), row.names = c(NA, -6L), class = c("data.table",
"data.frame"), .internal.selfref = <pointer: 0x0000014b25a51ef0>)
I'd like to create 7 bar plots (one for each "clade") where the X axis would have the columns of the data frame (C.T would be 1 bar, A.G would be another bar, etc) and the Y axis would be the count. Essentially, for each clade, print a barplot with the counts of column.
For example, for the bar plot of the clade "20B" and the bar name "C.T" the count would be the sum of the values from the data frame. Can I do that in this wide format? Do I need to transform the data to a long format instead?
I was trying to apply this SO answer: Plotting error bar on bar chart for a data frame in wide format using ggplot but I keep getting choose another strategy with names_repair
Thank you in advance, any help is very welcome!
I have a data set I'm trying to run a glm regression on, however it contains characters as age limit, race, and comorbidity class. I would like to change those columns into a continuous variable so the regression can accept it. Data below, I want to change the TBI.irace2 into (Hispanic=1, Black=2, white=3, and other=4) same with age (age 18-28=1, 29-46=2, 47-64=3, and >64=4) and with NISS (NISS 0-10=1, NISS 11-20=2, NISS 21-30=3, and NISS 31-40=4, NISS41-50=5, NISS 51-60=6, NISS 61-70=7, NISS>70= 8)
Please find summary of data below
TBI.crani = c(0, 0, 0, 0, 0, 0), TBI.vte = c(0,
0, 0, 0, 0, 0), TBI.FEMALE = c(0, 0, 1, 0, 1, 0), TBI.iracecat2 = c("Whites",
"Whites", "Whites", "Hispanics", "Whites", "Blacks"), TBI.agecat = c("Age 47-64",
"Age 29-46", "Age > 64", "Age 29-46", "Age 18-28", "Age 18-28"
), TBI.nisscategory = c("NISS 21-30", "NISS 11-20", "NISS 21-30",
"NISS 11-20", "NISS 11-20", "NISS 0-10"), TBI.LOS = c(5, 8, 1,
3, 19, 1), TBI.hospitalteach = c(0, 0, 1, 1, 1, 1), TBI.largebedsize = c(1,
1, 1, 1, 1, 1), TBI.CM_ALCOHOL = c(0, 0, 0, 1, 0, 0), TBI.CM_ANEMDEF = c(0,
0, 0, 0, 0, 0), TBI.CM_BLDLOSS = c(0, 0, 0, 0, 0, 0), TBI.CM_CHF = c(1,
0, 0, 0, 0, 0), TBI.CM_CHRNLUNG = c(0, 0, 0, 0, 0, 0), TBI.CM_COAG = c(0,
0, 0, 0, 1, 0), TBI.CM_HYPOTHY = c(0, 0, 0, 0, 0, 0), TBI.CM_LYTES = c(0,
0, 0, 0, 0, 0), TBI.CM_METS = c(0, 0, 0, 0, 0, 0), TBI.CM_NEURO = c(0,
0, 0, 0, 0, 0), TBI.CM_OBESE = c(0, 0, 0, 0, 0, 0), TBI.CM_PARA = c(0,
0, 0, 0, 0, 0), TBI.CM_PSYCH = c(0, 1, 0, 0, 0, 0), TBI.CM_TUMOR = c(0,
0, 0, 0, 0, 0), TBI.CM_WGHTLOSS = c(0, 0, 0, 0, 0, 0), TBI.UTI = c(0,
0, 0, 0, 0, 0), TBI.pneumonia = c(0, 0, 0, 0, 0, 0), TBI.AMI = c(0,
0, 0, 0, 0, 0), TBI.sepsis = c(0, 0, 0, 0, 0, 0), TBI.arrest = c(0,
0, 0, 0, 0, 0), TBI.spineinjury = c(0, 0, 0, 0, 0, 0), TBI.legfracture = c(0,
0, 0, 0, 0, 0), TBI_time_to_surg.NEW = c(0, 0, 0, 0, 0, 0)), row.names = c(NA,
6L), class = "data.frame")
A small little tip, provide a small sample set that is just big enough to address your question.
library(data.table)
# took a small sample and changed one value to Asian
dt <- data.table(
TBI.FEMALE = c(0, 0, 1, 0, 1, 0),
TBI.iracecat2 = as.character(c("Whites", "Whites", "Asian", "Hispanics", "Whites", "Blacks"))
)
# define race groups, and note I did not define Asian
convert_race <- c("Hispanics" = 1, "Blacks" = 2, "Whites" = 3) # other will all be not defined
dt[, TBI.irace2 := lapply(TBI.iracecat2, function(x) convert_race[x]), by = TBI.iracecat2]
dt[is.na(TBI.irace2), TBI.irace2 := 4]
dt
# TBI.FEMALE TBI.iracecat2 TBI.irace2
# 1: 0 Whites 3
# 2: 0 Whites 3
# 3: 1 Asian 4
# 4: 0 Hispanics 1
# 5: 1 Whites 3
# 6: 0 Blacks 2
I have two DF's:
passesComb <- structure(list(P1_Good = c(0, 1, 0, 0, 0, 0, 1), P2_Good = c(2,
0, 0, 0, 0, 0, 2), P3_Good = c(0, 1, 0, 0, 0, 0, 1), P4_Good = c(0,
0, 1, 0, 0, 0, 1), P5_Good = c(0, 0, 0, 1, 0, 0, 1), P1_Bad = c(0,
0, 0, 0, 0, 0, 0), P2_Bad = c(0, 0, 0, 0, 0, 0, 0), P3_Bad = c(0,
0, 0, 0, 0, 0, 0), P4_Bad = c(0, 0, 1, 0, 0, 0, 1), P5_Bad = c(0,
0, 0, 0, 0, 0, 0), `Bad Pass` = c(0, 0, 1, 0, 0, 1, 1), `Good Pass` = c(2,
2, 1, 1, 0, 3, 6), `Intercepted Pass` = c(0, 0, 0, 0, 0, 1, 0
), Turnover = c(0, 0, 0, 0, 0, 1, 0), totalEvents = c(2, 2, 2,
1, 0, 6, 7)), row.names = c("P1", "P2", "P3", "P4", "P5", "Opponent",
"VT"), class = "data.frame")
of size 7x15, and
copyComb <- structure(list(P1_Good = c(0, 1, 0, 0, 0, 1), P2_Good = c(2,
0, 0, 0, 0, 2), P4_Good = c(0, 0, 0, 0, 0, 0), P5_Good = c(0,
0, 1, 0, 0, 1), P1_Bad = c(0, 0, 0, 0, 0, 0), P2_Bad = c(0, 0,
0, 0, 0, 0), P3_Bad = c(0, 0, 0, 0, 0, 0), P4_Bad = c(0, 0, 0,
0, 0, 0), P5_Bad = c(0, 0, 0, 0, 0, 0), `Bad Pass` = c(0, 0,
0, 0, 1, 0), `Good Pass` = c(2, 1, 1, 0, 3, 4), `Intercepted Pass` = c(0,
0, 0, 0, 1, 0), Turnover = c(0, 0, 0, 0, 1, 0), totalEvents = c(2,
1, 1, 0, 6, 4)), row.names = c("P1", "P2", "P4", "P5", "Opponent",
"VT"), class = "data.frame")
or simply,
copyComb <- passesComb
copyComb <- copyComb[-3,-3]
#Updating specific cells since [3,3] is removed
copyComb[2,11] <- 1
copyComb[2,14] <- 1
copyComb[6,8] <- 0
copyComb[6,3] <- 0
copyComb[6,10] <- 0
copyComb[6,11] <- 4
copyComb[6,14] <- 4
#This now equals the copyComb from dput() above
of size 6x14.
I am trying to combine/add these two df's together based on matching row/column names. I tried to achieve this using the code from the answer to this post
gamesComb <- data.frame(matrix(NA, nrow = ifelse(nrow(passesComb) >= nrow(copyComb), nrow(passesComb),nrow(copyComb)),
ncol = ifelse(ncol(passesComb) >= ncol(copyComb), ncol(passesComb),ncol(copyComb))))
gamesComb[row.names(ifelse(nrow(passesComb) >= nrow(copyComb), passesComb, copyComb)),
colnames(ifelse(ncol(passesComb) >= ncol(copyComb), passesComb, copyComb))] <- passesComb
Here, I create a df, gamesComb and set the dimensions of whichever passesComb or copyComb is bigger. It does create a 7x15 df, but doesn't add the row/col names.
I also am trying to then add the 2 df's together based on the cell value if they have the same row/col name (same as in the post link above), i.e. passesComb["P2","P1_Good"] = 1 and copyComb["P2","P1_Good"] = 1, so gamesComb["P2","P1_Good"] should = 2, and same for all similar row/col names.
So the final result look like:
expectedOutput <- structure(list(P1_Good = c(0, 2, 0, 0, 0, 0, 2), P2_Good = c(4,
0, 0, 0, 0, 0, 4), P3_Good = c(0, 1, 0, 0, 0, 0, 1), P4_Good = c(0,
0, 1, 0, 0, 0, 1), P5_Good = c(0, 0, 0, 2, 0, 0, 2), P1_Bad = c(0,
0, 0, 0, 0, 0, 0), P2_Bad = c(0, 0, 0, 0, 0, 0, 0), P3_Bad = c(0,
0, 0, 0, 0, 0, 0), P4_Bad = c(0, 0, 1, 0, 0, 0, 1), P5_Bad = c(0,
0, 0, 0, 0, 0, 0), `Bad Pass` = c(0, 0, 1, 0, 0, 2, 1), `Good Pass` = c(4,
3, 1, 2, 0, 6, 10), `Intercepted Pass` = c(0, 0, 0, 0, 0, 2,
0), Turnover = c(0, 0, 0, 0, 0, 2, 0), totalEvents = c(4, 3,
2, 2, 0, 12, 11)), row.names = c("P1", "P2", "P3", "P4", "P5",
"Opponent", "VT"), class = "data.frame")
Here's a dplyr/tidyr approach where I reshape each table into a long format, then join them, sum, and pivot wider again.
library(dplyr); library(tidyr)
lengthen <- function(df) { df %>% rownames_to_column(var = "row") %>% pivot_longer(-row)}
full_join(lengthen(passesComb), lengthen(copyComb), by = c("row", "name")) %>%
mutate(new_val = coalesce(value.x, 0) + coalesce(value.y, 0)) %>%
select(-starts_with("value")) %>%
pivot_wider(names_from = name,values_from = new_val)
Another option is to stack them and then sum by rowname groups.
library(dplyr, warn.conflicts = FALSE)
library(tibble)
out <-
rownames_to_column(passesComb) %>%
bind_rows(rownames_to_column(copyComb)) %>%
# bind_rows(rownames_to_column(third_table)) %>% if you want to add another
select(rowname, names(passesComb)) %>%
group_by(rowname) %>%
summarise(across(everything(), sum, na.rm = T)) %>%
slice(match(rownames(passesComb), rowname)) %>%
column_to_rownames('rowname')
all.equal(out, expectedOutput)
#> [1] TRUE
Created on 2021-10-09 by the reprex package (v2.0.1)
I have been creating a bar plot for the result of a sentiment analysis model in R. The data is very confidential feedbacks from the customers. So, the feedbacks are then fed into a sentiment analysis model to generate outputs. My work is to generate a chart for each combination for example zone = delhi and delhi has sub zones like eastdelhi, westdelhi,northdelhi,southdelhi. I want to generate charts with combination like
zone = delhi and sub-zone = eastdelhi. And I want to save it to a jpeg file.I have written a for loop to do so. But for some reason it isn't working. This is the code
#Set locales
rm(list = ls())
Sys.setlocale(category = "LC_ALL",locale = "English")
#Load libraries
LoadLibraries <- c("openxlsx",
"dplyr",
"tidyr",
"plotly",
"RColorBrewer",
"shiny",
"officer",
"parallel",
"dplyr",
"tidyr",
"magrittr",
"knitr")
lapply(LoadLibraries, require, character.only = TRUE)
path = "C:/Users/R_Visual/Data/visual_data.xlsx"
input_data <- read.xlsx(path)
name <- names(input_data[,1:10])
#Filtering the zones and circles
for (i in 1:length(unique(Zone.Final))){
for (j in 1:length(unique(Circle.Final))){
fileName = 'C:/Users/R_Visual/'+ str(i) + str(j) + '.jpeg'
jpeg(fileName, width = 900, height = 450)
df <- input_data %>%
filter(input_data$Zone.Final[i])
df <- df %>%
filter(df$Circle.Final[j])
color <- c("#ca2f27","#f56d43","#f8c38a","#fde08b","#d9ef8b","#a7d86f","#67bd64","#1a984f","#D3D3D3","#A9A9A9")
plot <- barplot(sort(colSums(input_data[, 1:10])),
main = paste("Sentiment Analysis for Zone",df$Zone.Final[i]," and Circle",df$Circle.Final[j], sep = ""),
xlab = "Sentiments",
ylab = "Count",
horiz = FALSE,
names = name,
col = color,
border = FALSE,
legend = TRUE,
beside = TRUE,
legend.text = name,
args.legend = list(bty = "n", x = "topleft",ncol = 1, cex = 0.8, y.intersp = 0.8, x.intersp = 0.25, horiz = F, xpd = TRUE, inset = c(0,0)))
dev.off()
}
}
EDIT:
This is the sample of input_data
> dput(input_data)
structure(list(anger = c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0), anticipation = c(1,
0, 0, 0, 0, 0, 1, 0, 0, 0), disgust = c(0, 0, 0, 0, 0, 0, 0,
0, 0, 0), fear = c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0), joy = c(0,
0, 0, 0, 0, 0, 1, 0, 0, 0), sadness = c(0, 0, 0, 0, 0, 0, 0,
0, 0, 0), surprise = c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0), trust = c(0,
0, 1, 1, 1, 0, 2, 0, 0, 0), negative = c(0, 0, 0, 0, 0, 0, 0,
0, 0, 0), positive = c(1, 0, 0, 0, 1, 1, 2, 1, 0, 1), Zone.Final = c("Delhi",
"Lucknow", "Durgapur", "Lucknow", "Mumbai", "Bhopal", "Chandigarh",
"Chandigarh", "Gurugram", "Chandigarh"), Circle.Final = c("Noida",
"Gorakhpur", "Murshidabad", "Gorakhpur", "Mumbai City", "Bhopal",
"Chandigarh", "Panchkula", "Hisar", "Karnal")), row.names = c(NA,
10L), class = "data.frame")
If anyone could help me with the code, it would be of great help.
You can try creating a list combining the zone and subzone:
#Data
input_data <- structure(list(anger = c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0), anticipation = c(1,
0, 0, 0, 0, 0, 1, 0, 0, 0), disgust = c(0, 0, 0, 0, 0, 0, 0,
0, 0, 0), fear = c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0), joy = c(0,
0, 0, 0, 0, 0, 1, 0, 0, 0), sadness = c(0, 0, 0, 0, 0, 0, 0,
0, 0, 0), surprise = c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0), trust = c(0,
0, 1, 1, 1, 0, 2, 0, 0, 0), negative = c(0, 0, 0, 0, 0, 0, 0,
0, 0, 0), positive = c(1, 0, 0, 0, 1, 1, 2, 1, 0, 1), Zone.Final = c("Delhi",
"Lucknow", "Durgapur", "Lucknow", "Mumbai", "Bhopal", "Chandigarh",
"Chandigarh", "Gurugram", "Chandigarh"), Circle.Final = c("Noida",
"Gorakhpur", "Murshidabad", "Gorakhpur", "Mumbai City", "Bhopal",
"Chandigarh", "Panchkula", "Hisar", "Karnal")), row.names = c(NA,
10L), class = "data.frame")
#Code
#First create and global id to combine zone and subzone
df <- input_data
df$id <- paste(df$Zone.Final,df$Circle.Final,sep = '-')
#Split
List <- split(df,df$id)
#Plot
color <- c("#ca2f27","#f56d43","#f8c38a","#fde08b","#d9ef8b","#a7d86f","#67bd64","#1a984f","#D3D3D3","#A9A9A9")
#Plot names
vnames <- paste0(names(List),'.jpeg')
#Loop
for(i in 1:length(List))
{
name <- names(List[[i]][, 1:10])
#Plot
jpeg(filename = vnames[i], width = 900, height = 450)
barplot(sort(colSums(List[[i]][, 1:10])),
main = paste("Sentiment Analysis for Zone ",
unique(List[[i]]$Zone.Final),
" and Circle ",unique(List[[i]]$Circle.Final), sep = ""),
xlab = "Sentiments",
ylab = "Count",
horiz = FALSE,
names = name,
col = color,
border = FALSE,
legend = TRUE,
beside = TRUE,
legend.text = name,
args.legend = list(bty = "n", x = "topleft",ncol = 1,
cex = 0.8, y.intersp = 0.8, x.intersp = 0.25,
horiz = F, xpd = TRUE, inset = c(0,0)))
dev.off()
}
That will create the plots. Of course you can add a path to vnames like the dir you have to save the plots in that folder.