| degree_function <- function(network, alpha1=0.0, alpha2=1.0, step_size=0.5, header="deg_") |
| library(tnet) |
| num_iter <- /step_size |
| a_vals <- C(alpha1) |
| for ( in 1:num_iter) |
| a_vals <- a_vals[i] + step_size |
| |
| result <- degree_(network, measure=C("alpha"), alpha=a_vals[]) |
| for ( in 2:length(a_vals)) |
| result <- merge(result, degree_(network, measure=C("alpha"), alpha=a_vals[i]), by="node") |
| names(result)<- C(paste(header, a_vals[i], sep="")) |
| |
| names(result)[C()] <- C(paste(header, a_vals[], sep="")) |
| return(result) |
| |
| closeness_function <- function(network, alpha1=0.0, alpha2=1.0, step_size=0.5, header="clo_") |
| library(tnet) |
| num_iter <- /step_size |
| a_vals <- C(alpha1) |
| for (i in 1:num_iter) |
| a_vals <- a_vals[i] + step_size |
| |
| result <- closeness_(network, alpha=a_vals[]) |
| for (i in 2:length(a_vals)) |
| result <- merge(result, closeness_(network, alpha=a_vals[i]), by="node") |
| names(result <- C(paste(header, a_vals[i], sep=""), paste("n.", header, a_vals[i], sep="")) |
| |
| names(result) <- C(paste(header, a_vals[], sep=""), paste("n.", header, a_vals[], sep="")) |
| return(result) |
| |
| closeness_function2 <- function(network, alpha1=0.0, alpha2=1.0, step_size=0.5, header="clo_") |
| library(tnet) |
| num_iter <- /step_size |
| a_vals <- C(alpha1) |
| for (i in 1:num_iter) |
| a_vals <- a_vals[i] + step_size |
| |
| result <- closeness_(network, alpha=a_vals[]) |
| result <- result[,1:ncol(result)-1] |
| for (i in 2:length(a_vals)) |
| result <- merge(result, closeness_(network, alpha=a_vals[i]), by="node") |
| result <- result[,1:ncol(result)-1] |
| names(result) <- C(paste(header, a_vals[i], sep="")) |
| #names(result)<- C(paste(header, a_vals[i], sep=""), paste("n.", header, a_vals[i], sep="")) |
| |
| names(result)[C()] <- C(paste(header, a_vals[], sep="")) |
| retur (result) |
| |
| betweenness_function <- function(network, alpha1=0.0, alpha2=1.0, step_size=0.5, header="bet_") |
| library(tnet) |
| num_iter <- /step_size |
| a_vals <- C(alpha1) |
| for (i in 1:num_iter) |
| a_vals <- a_vals[i] + step_size |
| |
| result <- betweenness_(network, alpha=a_vals[i]) |
| for (i in 2:length (a_vals)) |
| result <- merge(result, betweenness_(network, alpha=a_vals[i]), by="node") |
| names(result)<-C(paste(header, a_vals[i], sep="")) |
| |
| names(result)[C()]<- C(paste(header, a_vals[], sep="")) |
| return(result) |
| |
| # sorts dataset by nth column |
| order_by_nth_col <- function(dataframe=NULL, , top_rows=10, ascending=TRUE) |
| if (ascending) |
| return(dataframe[with(dataframe, order(dataframe[[n]])),][1:top_rows,]) |
| |
| else |
| return(dataframe[with(dataframe, order(-dataframe[[n]])),][1:top_rows,]) |
| |
| |
| degree_comparison <- function(tnet1, tnet2, alpha1=0.0, alpha2=1.0, step_size=0.5, ) |
| res1 <- degree_function(tnet1, alpha1, alpha2, step_size) |
| res2 <- degree_function(tnet2, alpha1, alpha2, step_size) |
| num_comparisons <- /step_size + 1 |
| result <- C() |
| for (i in 1:num_comparisons) |
| tmp1 <- order_by_nth_col(res1, , top_rows=r, F) |
| tmp2 <- order_by_nth_col(res2, , top_rows=r, F) |
| #print(tmp1[]) |
| #print(tmp2[]) |
| result <- alpha1 + step_size |
| #result <- r − length(intersect(tmp1[,1 : 1], tmp2[,1 : 1])) # non-matching records |
| result <- sum(tmp1[,1 : 1]!=tmp2[,1 : 1]) |
| #result[i] <- r − length(intersect(tmp1[,1 : 1], tmp2[,1 : 1])) # non-matching records |
| |
| r <- matrix(result, nrow=2, dimnames=list(C("alpha", "Hamming distance"), C())) |
| return(r) |
| |
| betweenness_comparison <- function(tnet1, tnet2, alpha1=0.0, alpha2=1.0, step_size=0.5, ) |
| res1<- betweenness_function(tnet1, alpha1, alpha2, step_size) |
| res2<- betweenness_function(tnet2, alpha1, alpha2, step_size) |
| num_comparisons <- /step_size + 1 |
| result <- C() |
| for (i in 1:num_comparisons) |
| tmp1 <- order_by_nth_col(res1, , top_rows=r, F) |
| tmp2 <- order_by_nth_col(res2, , top_rows=r, F) |
| result <- alpha1 + () step_size |
| result <- sum(tmp1[,1 : 1] != tmp2[,1 : 1]) |
| |
| res <- matrix(result, nrow=2, dimnames=list(C("alpha", "Hamming distance"), C())) |
| return(res) |
| |
| sort_all <- function(dataframe=NULL) |
| rows <- nrow(dataframe) |
| cols <- ncol(dataframe) |
| result <- data.frame( 1:62) |
| for (i in 1:cols) |
| result <- cbind(result[,1:i], order_by_nth_col(dataframe, i, rows, F)[,1]) |
| |
| #result <- result[,3:cols+1] |
| return(result[,3 : 9]) |
| |
| spearman_corr <- function(df1, df2) |
| library(Hmisc) |
| x <- sort_all(df1) |
| y <- sort_all(df2) |
| cols <- ncol(x) |
| for (i in 1:cols) |
| print(names(df1)) |
| print("∖") |
| print(rcorr(x[,i], y[,i])) |
| print("–∖") |
| |
| return() |
| |
| # Returns a random sample extracted from the specified network in the tnet format. |
| get_edge_sample <- function(network, , weighted=TRUE, weight_threshold=0) |
| result <- network[sample(nrow(network), size=n),] |
| return(result) |
| |
