Hey, are you looking for this? statisticsglobe.com/convert-row-names-into-column-of-data-frame-in-r

Hey, I haven’t done any direct comparisons, but I believe using solve() for this purpose is standard practice in R.

Thanks! I recommend using the FactoMineR package for correspondence analysis. It’s robust, widely used, and works seamlessly with factoextra for visualizations.

Hey, could you please share your code and the structure of your data?

@@StatisticsGlobesure can i send you via email?

It would be great if you could post it here, so that others can read/contribute as well. Thanks!

@@StatisticsGlobe this is my input.R file # Prepare dataset as input to Klink-2. data_dir <- "../../data/klink2/" source("relations.R") Rcpp::sourceCpp("utils.cpp") # input relations taken into consideration relations <- c("publication", "author", "venue", "area") # number of relations rn <- length(relations) # length of saved cooccurrence values m <- 100 # 4 input variables to Klink-2: keywordsdb <- new.env(parent = globalenv(), hash = TRUE) reldb_df <- list() reldb_l <- list() inputm <- matrix(, nrow = m, ncol = 0) # reads dataset into global raw input variables: keywordsdb, reldb_df, reldb_l read_dataset <- function(limit = -1, named_list) { file_name <- paste(data_dir, named_list, "/", named_list, ".tsv", sep = "") d <- read.csv( file_name, sep = "\t", header = TRUE, stringsAsFactors = FALSE, nrows = limit * 2 ) empty <- function(elem) is.null(elem) || elem == "" || all(trimws(elem) == "") # need only items with defined keywords fields (DE, ID) w <- c() for (i in 1:dim(d)[1]) { if (!(empty(d$DE[i]) && empty(d$ID[i]))) w <- c(i, w) } d <- d[w, ] if (limit > 0) d <- d[1:limit, ] # fields used: document title, authors, publication name, research areas, year d <- d[c("DE", "TI", "AU", "SO", "SC", "PY")] k <- 1 a <- 1 n <- nrow(d) process_item <- function(item) { cat("process article ", a, " out of ", n, " ") a <<- a + 1 # Process keywords with error handling keywords <- tryCatch({ unique(unlist( sapply(strsplit(tolower(item["DE"]), ";"), function(x) { x <- trimws(x) x[nzchar(x)] # Only keep non-zero length strings }) )) }, error = function(e) character(0)) # Skip if no valid keywords if (length(keywords) == 0 || all(keywords == "")) return() newkeywords <- setdiff(keywords, ls(keywordsdb)) oldkeywords <- setdiff(keywords, newkeywords) # Process other fields with error handling authors <- tryCatch({ res <- unlist(sapply(strsplit(item["AU"], ";"), function(x) trimws(x[nzchar(x)]))) if (is.null(res)) character(0) else as.vector(res) }, error = function(e) character(0)) areas <- tryCatch({ res <- unique(unlist(sapply(strsplit(tolower(item["SC"]), ";"), function(x) trimws(x[nzchar(x)])))) if (is.null(res)) character(0) else res }, error = function(e) character(0)) venues <- if (!empty(item["SO"])) tolower(trimws(item["SO"])) else character(0) # Create relation vectors only for non-empty elements relation <- c(1, rep(2, length(authors)), 3, rep(4, length(areas))) entity <- c(item["TI"], authors, venues, areas) quantity <- rep(NA_integer_, length(entity)) year <- as.numeric(rep(item["PY"], length(entity))) # Process new keywords for (i in seq_along(newkeywords)) { if (nzchar(newkeywords[i])) { # Only process non-empty keywords keywordsdb[[newkeywords[i]]] <<- k reldb_l[[k]] <<- list() reldb_l[[k]]$publication <<- paste(item["TI"], year[1], sep = "_") reldb_l[[k]]$author <<- if(length(authors) > 0) vapply(authors, paste, "", year[1], sep = "_") else character(0) reldb_l[[k]]$venue <<- if(length(venues) > 0) paste(venues, year[1], sep = "_") else character(0) reldb_l[[k]]$area <<- if(length(areas) > 0) vapply(areas, paste, "", year[1], sep = "_") else character(0) names(reldb_l)[k] <<- newkeywords[i] reldb_df[[k]] <<- data.frame(relation, entity, quantity, year, stringsAsFactors = FALSE) names(reldb_df)[k] <<- newkeywords[i] k <<- k + 1 } } # Process existing keywords for (i in seq_along(oldkeywords)) { if (nzchar(oldkeywords[i])) { index <- keywordsdb[[oldkeywords[i]]] reldb_l[[index]]$publication <<- c(reldb_l[[index]]$publication, paste(item["TI"], year[1], sep = "_")) reldb_l[[index]]$author <<- unique(c(reldb_l[[index]]$author, if(length(authors) > 0) vapply(authors, paste, "", year[1], sep = "_") else character(0))) reldb_l[[index]]$venue <<- unique(c(reldb_l[[index]]$venue, if(length(venues) > 0) paste(venues, year[1], sep = "_") else character(0))) reldb_l[[index]]$area <<- unique(c(reldb_l[[index]]$area, if(length(areas) > 0) vapply(areas, paste, "", year[1], sep = "_") else character(0))) reldb_df[[index]] <<- rbind(reldb_df[[index]], data.frame(relation, entity, quantity, year, stringsAsFactors = FALSE)) } } } apply(d, 1, process_item) # Sort the fields in reldb_l for (i in 1:length(reldb_l)) { if (!is.null(reldb_l[[i]]$publication)) reldb_l[[i]]$publication <<- sort(reldb_l[[i]]$publication) if (!is.null(reldb_l[[i]]$author)) reldb_l[[i]]$author <<- sort(reldb_l[[i]]$author) if (!is.null(reldb_l[[i]]$venue)) reldb_l[[i]]$venue <<- sort(reldb_l[[i]]$venue) if (!is.null(reldb_l[[i]]$area)) reldb_l[[i]]$area <<- sort(reldb_l[[i]]$area) } } # Rest of the functions remain unchanged entities_range <- function(reldb_l) { a <- Inf b <- 0 for (i in 1:length(names(reldb_l))) { for (r in 1:rn) { t <- length(reldb_l[[i]][[r]]) if (t < a) a <- t if (t > b) b <- t } } c(a, b) } cache_cooccurrence <- function() { n <- length(ls(keywordsdb)) inputm <- matrix(0, nrow = m, ncol = n * 2 * rn) maxsize <- entities_range(reldb_l)[2] for (i in 1:n) { cat("process keyword ", i, " out of ", n, " ") irel <- reldb_l[[i]] for (r in 1:rn) { co_m <- calc_cooccurrence_C(n, m, i, r, reldb_l, maxsize) inputm[, cached.keys(i, r)] <- co_m[, 1] inputm[, cached.values(i, r)] <- co_m[, 2] } } inputm } run_all <- function(limit = -1, named_list) { keywordsdb <<- new.env(parent = globalenv(), hash = TRUE) reldb_df <<- list() reldb_l <<- list() read_dataset(limit, named_list) inputm <<- cache_cooccurrence() if (limit > 0) { fname <- paste(data_dir, named_list, "/", named_list, limit, ".Rdata", sep = "") } else { fname <- paste(data_dir, named_list, "/", named_list, ".Rdata", sep = "") } save("reldb_df", "reldb_l", "keywordsdb", "inputm", file = fname) cat("Input variables saved to", fname, " ") } inspect_dataset <- function(filename) { file_name <- paste(data_dir, filename, "/", filename, ".Rdata", sep = "") load(file_name) n <- length(reldb_df) entrange <- entities_range(reldb_l) cat("Number of keywords: ", n, " ") #cat("Range of entities associated with a keyword: [", entrange[1], ",", entrange[2], "] ") for (i in 1:rn) { cat( "Value of co-occurrence,", relations[i], "relation: [", min(inputm[, cached.values(1:n, i)]), ",", max(inputm[, cached.values(1:n, i)]), "] " ) } }

i got above error on this step¨Run the Klink-2 Scripts¨

Hey, I run a Facebook discussion group where people can ask questions about R programming and statistics. Could you post your question there? This way, others can contribute/read as well: facebook.com/groups/statisticsglobe For more detailed, personalized assistance, you may also take a look at our consulting page: statisticsglobe.com/consulting

Hey, how about this? library(ggplot2) # Example data data <- data.frame( week = 1:13, value = rnorm(13) ) # Custom labels with line breaks ggplot(data, aes(x = week, y = value)) + geom_line() + geom_point() + scale_x_continuous( breaks = 1:13, labels = c("1 Run-in", "2", "3", "4", "5", "6", "7 Supplement", "8", "9", "10", "11", "12", "13 Washout") )

Hey, you will need to define a custom function for this. The time2season function uses default boundaries for seasons, but it doesn’t allow you to directly modify those defaults. Instead, you can create your own logic for assigning seasons based on your country’s calendar. Here’s how you can do it: # Define example dates my_dates <- as.Date(c("2022-10-01", "2021-05-13", "2025-12-01", "2023-02-17", "2023-06-25", "2022-10-15")) my_dates # Print example dates # Custom function to assign seasons assign_season <- function(dates) { # Extract month and day month <- as.numeric(format(dates, "%m")) day <- as.numeric(format(dates, "%d")) # Assign seasons based on your custom boundaries ifelse((month == 12 & day >= 21) | month %in% c(1, 2) | (month == 3 & day < 21), "Winter", ifelse((month == 3 & day >= 21) | month %in% c(4, 5) | (month == 6 & day < 21), "Spring", ifelse((month == 6 & day >= 21) | month %in% c(7, 8) | (month == 9 & day < 21), "Summer", "Autumn"))) } # Apply the custom function my_seasons <- assign_season(my_dates) my_seasons # Print custom seasons

Thanks! Both approaches are possible. You can test the assumptions either before or after performing ANOVA, as long as you ensure that the assumption tests are completed and any violations addressed before interpreting or sharing your results.

Thank you for the kind comment and your feedback!

Thanks a lot for your kind feedback. Glad it helped! 🙂

Thank you very much for your kind feedback! 🙂

You are welcome, glad it was helpful!

Hey, one solution could be to call geom_area multiple times.

You're most welcome. Glad it helped! 🙂

Thanks for sharing your insights!

Thank you very much for your kind feedback! 🙂

Thanks a lot for the kind feedback! :)

Thanks, glad you like it!

You're most welcome! 🙂

Thanks a lot for your kind feedback. Glad it helped! 🙂

Hey, for data frames with many variables containing outliers, it’s best to address outliers carefully, often one variable at a time. Rather than removing them outright, consider retaining all data and applying appropriate statistical methods to handle outliers, as they may hold valuable insights or represent unique cases.

Haha, indeed that's something you have to get used to.

That's surprising. I just ran the code again and for me it works fine.

Thanks for the feedback. Indeed, this is a complex topic. Let me know if you have any specific questions.

Thank you very much for your kind feedback! 🙂

Thank you so much for the kind comment and for sharing your code!

You're most welcome. Glad it helped! 🙂

You're most welcome! 🙂

Thanks a lot for your kind feedback. Glad it helped! 🙂

Thank you so much for the very kind feedback, this is great to hear!

You're most welcome! 🙂

Thank you for the kind comment and hint!

Hey, thank you so much for the kind comment and topic suggestion. I'll consider it for a future video.

Hey, could you share your code?

You're most welcome! 🙂

Hey, you may use the following code: rbind(new_data, old_data)

Thank you very much for your kind feedback! 🙂

Thank you for the kind comment and the tip! It appears this method works as well. I'm not sure why I did it differently in the video, but perhaps this functionality was added more recently. The video is already over three years old.

You're most welcome! 🙂

Thanks, glad you like it! :)

Hey, did you run the code exactly as demonstrated in the video?

@@StatisticsGlobe Yes as it is

That's weird, to be honest, I don't know why this is happening. On my side, everything works fine.

Please see my response to your other question.

Sehr gerne! 🙂