diff --git a/.Rhistory b/.Rhistory
index 3dc0977..6d2283c 100644
--- a/.Rhistory
+++ b/.Rhistory
@@ -1,512 +1,512 @@
-g.breaks.clean <- g.breaks[nz.probs.mask]
-g.probs.clean <- g.breaks[nz.probs.mask]
-#plot(log(g.breaks.clean), log(g.probs.clean))
-g.fit <- lm(log(g.probs.clean)~log(g.breaks.clean))
-summary(g.fit)
-alpha.LM <- coef(g.fit)[2]
-lines(g.seq, g.seq^(-alpha.LM), col="#E66100", lty=3)
-################# Max-Log-Likelihood #################
-n <- length(g.breaks.clean)
-kmin <- g.breaks.clean[1]
-alpha.ML <- 1 + n/sum(log(g.breaks.clean)/kmin)
-alpha.ML
-lines(g.seq, g.seq^(-alpha.ML), col="#D35FB7", lty=4)
-# Homework 4 for the University of Tulsa' s CS-7863 Network Theory Course
-# Degree Distribution
-# Professor: Dr. McKinney, Spring 2022
-# Noah Schrick - 1492657
-library(igraph)
-library(igraphdata)
-data(yeast)
-g <- yeast
-g.netname <- "Yeast"
-################# Set up Work #################
-g.vec <- degree(g)
-g.hist <- hist(g.vec, freq=FALSE, main=paste("Histogram of the", g.netname,
-" Network"))
-legend("topright", c("Guess", "Poisson", "Least-Squares Fit",
-"Max Log-Likelihood"), lty=c(1,2,3,4), col=c("#40B0A6",
-"#006CD1", "#E66100", "#D35FB7"))
-g.mean <- mean(g.vec)
-g.seq <- 0:max(g.vec) # x-axis
-################# Guessing Alpha #################
-alpha.guess <- 1.5
-lines(g.seq, g.seq^(-alpha.guess), col="#40B0A6", lty=1)
-################# Poisson #################
-g.pois <- dpois(g.seq, g.mean, log=F)
-lines(g.seq, g.pois, col="#006CD1", lty=2)
-################# Linear model: Least-Squares Fit #################
-g.breaks <- g.hist$breaks[-c(1)] # remove 0
-g.probs <- g.hist$density[-1] # make lengths match
-# Need to clean up probabilities that are 0
-nz.probs.mask <- g.probs!=0
-g.breaks.clean <- g.breaks[nz.probs.mask]
-g.probs.clean <- g.breaks[nz.probs.mask]
-#plot(log(g.breaks.clean), log(g.probs.clean))
-g.fit <- lm(log(g.probs.clean)~log(g.breaks.clean))
-summary(g.fit)
-alpha.LM <- coef(g.fit)[2]
-lines(g.seq, g.seq^(-alpha.LM), col="#E66100", lty=3)
-################# Max-Log-Likelihood #################
-n <- length(g.breaks.clean)
-kmin <- g.breaks.clean[1]
-alpha.ML <- 1 + n/sum(log(g.breaks.clean)/kmin)
-alpha.ML
-lines(g.seq, g.seq^(-alpha.ML), col="#D35FB7", lty=4)
-# Homework 4 for the University of Tulsa' s CS-7863 Network Theory Course
-# Degree Distribution
-# Professor: Dr. McKinney, Spring 2022
-# Noah Schrick - 1492657
-library(igraph)
-library(igraphdata)
-data(yeast)
-g <- yeast
-g.netname <- "Yeast"
-################# Set up Work #################
-g.vec <- degree(g)
-g.hist <- hist(g.vec, freq=FALSE, main=paste("Histogram of the", g.netname,
-" Network"))
-legend("topright", c("Guess", "Poisson", "Least-Squares Fit",
-"Max Log-Likelihood"), lty=c(1,2,3,4), col=c("#40B0A6",
-"#006CD1", "#E66100", "#D35FB7"))
-g.mean <- mean(g.vec)
-g.seq <- 0:max(g.vec) # x-axis
-################# Guessing Alpha #################
-alpha.guess <- 1.5
-lines(g.seq, g.seq^(-alpha.guess), col="#40B0A6", lty=1)
-################# Poisson #################
-g.pois <- dpois(g.seq, g.mean, log=F)
-lines(g.seq, g.pois, col="#006CD1", lty=2)
-################# Linear model: Least-Squares Fit #################
-g.breaks <- g.hist$breaks[-c(1,2)] # remove 0
-g.probs <- g.hist$density[-1] # make lengths match
-# Need to clean up probabilities that are 0
-nz.probs.mask <- g.probs!=0
-g.breaks.clean <- g.breaks[nz.probs.mask]
-g.probs.clean <- g.breaks[nz.probs.mask]
-#plot(log(g.breaks.clean), log(g.probs.clean))
-g.fit <- lm(log(g.probs.clean)~log(g.breaks.clean))
-summary(g.fit)
-alpha.LM <- coef(g.fit)[2]
-lines(g.seq, g.seq^(-alpha.LM), col="#E66100", lty=3)
-################# Max-Log-Likelihood #################
-n <- length(g.breaks.clean)
-kmin <- g.breaks.clean[1]
-alpha.ML <- 1 + n/sum(log(g.breaks.clean)/kmin)
-alpha.ML
-lines(g.seq, g.seq^(-alpha.ML), col="#D35FB7", lty=4)
-# Homework 4 for the University of Tulsa' s CS-7863 Network Theory Course
-# Degree Distribution
-# Professor: Dr. McKinney, Spring 2022
-# Noah Schrick - 1492657
-library(igraph)
-library(igraphdata)
-data(yeast)
-g <- yeast
-g.netname <- "Yeast"
-################# Set up Work #################
-g.vec <- degree(g)
-g.hist <- hist(g.vec, freq=FALSE, main=paste("Histogram of the", g.netname,
-" Network"))
-legend("topright", c("Guess", "Poisson", "Least-Squares Fit",
-"Max Log-Likelihood"), lty=c(1,2,3,4), col=c("#40B0A6",
-"#006CD1", "#E66100", "#D35FB7"))
-g.mean <- mean(g.vec)
-g.seq <- 0:max(g.vec) # x-axis
-################# Guessing Alpha #################
-alpha.guess <- 1.5
-lines(g.seq, g.seq^(-alpha.guess), col="#40B0A6", lty=1)
-################# Poisson #################
-g.pois <- dpois(g.seq, g.mean, log=F)
-lines(g.seq, g.pois, col="#006CD1", lty=2)
-################# Linear model: Least-Squares Fit #################
-g.breaks <- g.hist$breaks[-c(1,2,3)] # remove 0
-g.probs <- g.hist$density[-1] # make lengths match
-# Need to clean up probabilities that are 0
-nz.probs.mask <- g.probs!=0
-g.breaks.clean <- g.breaks[nz.probs.mask]
-g.probs.clean <- g.breaks[nz.probs.mask]
-#plot(log(g.breaks.clean), log(g.probs.clean))
-g.fit <- lm(log(g.probs.clean)~log(g.breaks.clean))
-summary(g.fit)
-alpha.LM <- coef(g.fit)[2]
-lines(g.seq, g.seq^(-alpha.LM), col="#E66100", lty=3)
-################# Max-Log-Likelihood #################
-n <- length(g.breaks.clean)
-kmin <- g.breaks.clean[1]
-alpha.ML <- 1 + n/sum(log(g.breaks.clean)/kmin)
-alpha.ML
-lines(g.seq, g.seq^(-alpha.ML), col="#D35FB7", lty=4)
-# Homework 4 for the University of Tulsa' s CS-7863 Network Theory Course
-# Degree Distribution
-# Professor: Dr. McKinney, Spring 2022
-# Noah Schrick - 1492657
-library(igraph)
-library(igraphdata)
-data(yeast)
-g <- yeast
-g.netname <- "Yeast"
-################# Set up Work #################
-g.vec <- degree(g)
-g.hist <- hist(g.vec, freq=FALSE, main=paste("Histogram of the", g.netname,
-" Network"))
-legend("topright", c("Guess", "Poisson", "Least-Squares Fit",
-"Max Log-Likelihood"), lty=c(1,2,3,4), col=c("#40B0A6",
-"#006CD1", "#E66100", "#D35FB7"))
-g.mean <- mean(g.vec)
-g.seq <- 0:max(g.vec) # x-axis
-################# Guessing Alpha #################
-alpha.guess <- 1.5
-lines(g.seq, g.seq^(-alpha.guess), col="#40B0A6", lty=1)
-################# Poisson #################
-g.pois <- dpois(g.seq, g.mean, log=F)
-lines(g.seq, g.pois, col="#006CD1", lty=2)
-################# Linear model: Least-Squares Fit #################
-g.breaks <- g.hist$breaks[-c(1)] # remove 0
-g.probs <- g.hist$density[-1] # make lengths match
-# Need to clean up probabilities that are 0
-nz.probs.mask <- g.probs!=0
-g.breaks.clean <- g.breaks[nz.probs.mask]
-g.probs.clean <- g.breaks[nz.probs.mask]
-#plot(log(g.breaks.clean), log(g.probs.clean))
-g.fit <- lm(log(g.probs.clean)~log(g.breaks.clean))
-summary(g.fit)
-alpha.LM <- coef(g.fit)[2]
-lines(g.seq, g.seq^(-alpha.LM), col="#E66100", lty=3)
-################# Max-Log-Likelihood #################
-n <- length(g.breaks.clean)
-kmin <- g.breaks.clean[1]
-alpha.ML <- 1 + n/sum(log(g.breaks.clean)/kmin)
-alpha.ML
-lines(g.seq, g.seq^(-alpha.ML), col="#D35FB7", lty=4)
-# Homework 4 for the University of Tulsa' s CS-7863 Network Theory Course
-# Degree Distribution
-# Professor: Dr. McKinney, Spring 2022
-# Noah Schrick - 1492657
-library(igraph)
-library(igraphdata)
-data(yeast)
-g <- yeast
-g.netname <- "Yeast"
-################# Set up Work #################
-g.vec <- degree(g)
-g.hist <- hist(g.vec, freq=FALSE, main=paste("Histogram of the", g.netname,
-" Network"))
-legend("topright", c("Guess", "Poisson", "Least-Squares Fit",
-"Max Log-Likelihood"), lty=c(1,2,3,4), col=c("#40B0A6",
-"#006CD1", "#E66100", "#D35FB7"))
-g.mean <- mean(g.vec)
-g.seq <- 0:max(g.vec) # x-axis
-################# Guessing Alpha #################
-alpha.guess <- 1.5
-lines(g.seq, g.seq^(-alpha.guess), col="#40B0A6", lty=1)
-################# Poisson #################
-g.pois <- dpois(g.seq, g.mean, log=F)
-lines(g.seq, g.pois, col="#006CD1", lty=2)
-################# Linear model: Least-Squares Fit #################
-#g.breaks <- g.hist$breaks[-c(1)] # remove 0
-g.breaks <- g.hist$breaks # remove 0
-g.probs <- g.hist$density[-1] # make lengths match
-# Need to clean up probabilities that are 0
-nz.probs.mask <- g.probs!=0
-g.breaks.clean <- g.breaks[nz.probs.mask]
-g.probs.clean <- g.breaks[nz.probs.mask]
-#plot(log(g.breaks.clean), log(g.probs.clean))
-g.fit <- lm(log(g.probs.clean)~log(g.breaks.clean))
-summary(g.fit)
-alpha.LM <- coef(g.fit)[2]
-lines(g.seq, g.seq^(-alpha.LM), col="#E66100", lty=3)
-################# Max-Log-Likelihood #################
-n <- length(g.breaks.clean)
-kmin <- g.breaks.clean[1]
-alpha.ML <- 1 + n/sum(log(g.breaks.clean)/kmin)
-alpha.ML
-lines(g.seq, g.seq^(-alpha.ML), col="#D35FB7", lty=4)
-# Homework 4 for the University of Tulsa' s CS-7863 Network Theory Course
-# Degree Distribution
-# Professor: Dr. McKinney, Spring 2022
-# Noah Schrick - 1492657
-library(igraph)
-library(igraphdata)
-data(yeast)
-g <- yeast
-g.netname <- "Yeast"
-################# Set up Work #################
-g.vec <- degree(g)
-g.hist <- hist(g.vec, freq=FALSE, main=paste("Histogram of the", g.netname,
-" Network"))
-legend("topright", c("Guess", "Poisson", "Least-Squares Fit",
-"Max Log-Likelihood"), lty=c(1,2,3,4), col=c("#40B0A6",
-"#006CD1", "#E66100", "#D35FB7"))
-g.mean <- mean(g.vec)
-g.seq <- 0:max(g.vec) # x-axis
-################# Guessing Alpha #################
-alpha.guess <- 1.5
-lines(g.seq, g.seq^(-alpha.guess), col="#40B0A6", lty=1)
-################# Poisson #################
-g.pois <- dpois(g.seq, g.mean, log=F)
-lines(g.seq, g.pois, col="#006CD1", lty=2)
-################# Linear model: Least-Squares Fit #################
-g.breaks <- g.hist$breaks[-c(1)] # remove 0
-g.probs <- g.hist$density[-1] # make lengths match
-# Need to clean up probabilities that are 0
-nz.probs.mask <- g.probs!=0
-g.breaks.clean <- g.breaks[nz.probs.mask]
-g.probs.clean <- g.probs[nz.probs.mask]
-#plot(log(g.breaks.clean), log(g.probs.clean))
-g.fit <- lm(log(g.probs.clean)~log(g.breaks.clean))
-summary(g.fit)
-alpha.LM <- coef(g.fit)[2]
-lines(g.seq, g.seq^(-alpha.LM), col="#E66100", lty=3)
-################# Max-Log-Likelihood #################
-n <- length(g.breaks.clean)
-kmin <- g.breaks.clean[1]
-alpha.ML <- 1 + n/sum(log(g.breaks.clean)/kmin)
-alpha.ML
-lines(g.seq, g.seq^(-alpha.ML), col="#D35FB7", lty=4)
-alpha.LM
-# Homework 4 for the University of Tulsa' s CS-7863 Network Theory Course
-# Degree Distribution
-# Professor: Dr. McKinney, Spring 2022
-# Noah Schrick - 1492657
-library(igraph)
-library(igraphdata)
-data(yeast)
-g <- yeast
-g.netname <- "Yeast"
-################# Set up Work #################
-g.vec <- degree(g)
-g.hist <- hist(g.vec, freq=FALSE, main=paste("Histogram of the", g.netname,
-" Network"))
-legend("topright", c("Guess", "Poisson", "Least-Squares Fit",
-"Max Log-Likelihood"), lty=c(1,2,3,4), col=c("#40B0A6",
-"#006CD1", "#E66100", "#D35FB7"))
-g.mean <- mean(g.vec)
-g.seq <- 0:max(g.vec) # x-axis
-################# Guessing Alpha #################
-alpha.guess <- 1.5
-lines(g.seq, g.seq^(-alpha.guess), col="#40B0A6", lty=1)
-################# Poisson #################
-g.pois <- dpois(g.seq, g.mean, log=F)
-lines(g.seq, g.pois, col="#006CD1", lty=2)
-################# Linear model: Least-Squares Fit #################
-g.breaks <- g.hist$breaks[-c(1)] # remove 0
-g.probs <- g.hist$density[-1] # make lengths match
-# Need to clean up probabilities that are 0
-nz.probs.mask <- g.probs!=0
-g.breaks.clean <- g.breaks[nz.probs.mask]
-g.probs.clean <- g.probs[nz.probs.mask]
-#plot(log(g.breaks.clean), log(g.probs.clean))
-g.fit <- lm(log(g.probs.clean)~log(g.breaks.clean))
-summary(g.fit)
-alpha.LM <- coef(g.fit)[2]
-lines(g.seq, g.seq^(-alpha.LM), col="#E66100", lty=3)
-################# Max-Log-Likelihood #################
-n <- length(g.breaks.clean)
-kmin <- g.breaks.clean[1]
-alpha.ML <- 1 + n/sum(log(g.breaks.clean/kmin))
-alpha.ML
-lines(g.seq, g.seq^(-alpha.ML), col="#D35FB7", lty=4)
-# Homework 4 for the University of Tulsa' s CS-7863 Network Theory Course
-# Degree Distribution
-# Professor: Dr. McKinney, Spring 2022
-# Noah Schrick - 1492657
-library(igraph)
-library(igraphdata)
-data(yeast)
-g <- yeast
-g.netname <- "Yeast"
-################# Set up Work #################
-g.vec <- degree(g)
-g.hist <- hist(g.vec, freq=FALSE, main=paste("Histogram of the", g.netname,
-" Network"))
-legend("topright", c("Guess", "Poisson", "Least-Squares Fit",
-"Max Log-Likelihood"), lty=c(1,2,3,4), col=c("#40B0A6",
-"#006CD1", "#E66100", "#D35FB7"))
-g.mean <- mean(g.vec)
-g.seq <- 0:max(g.vec) # x-axis
-################# Guessing Alpha #################
-alpha.guess <- 1.5
-lines(g.seq, g.seq^(-alpha.guess), col="#40B0A6", lty=1, lwd=5)
-################# Poisson #################
-g.pois <- dpois(g.seq, g.mean, log=F)
-lines(g.seq, g.pois, col="#006CD1", lty=2)
-################# Linear model: Least-Squares Fit #################
-g.breaks <- g.hist$breaks[-c(1)] # remove 0
-g.probs <- g.hist$density[-1] # make lengths match
-# Need to clean up probabilities that are 0
-nz.probs.mask <- g.probs!=0
-g.breaks.clean <- g.breaks[nz.probs.mask]
-g.probs.clean <- g.probs[nz.probs.mask]
-#plot(log(g.breaks.clean), log(g.probs.clean))
-g.fit <- lm(log(g.probs.clean)~log(g.breaks.clean))
-summary(g.fit)
-alpha.LM <- coef(g.fit)[2]
-lines(g.seq, g.seq^(-alpha.LM), col="#E66100", lty=3)
-################# Max-Log-Likelihood #################
-n <- length(g.breaks.clean)
-kmin <- g.breaks.clean[1]
-alpha.ML <- 1 + n/sum(log(g.breaks.clean/kmin))
-alpha.ML
-lines(g.seq, g.seq^(-alpha.ML), col="#D35FB7", lty=4)
-# Homework 4 for the University of Tulsa' s CS-7863 Network Theory Course
-# Degree Distribution
-# Professor: Dr. McKinney, Spring 2022
-# Noah Schrick - 1492657
-library(igraph)
-library(igraphdata)
-data(yeast)
-g <- yeast
-g.netname <- "Yeast"
-################# Set up Work #################
-g.vec <- degree(g)
-g.hist <- hist(g.vec, freq=FALSE, main=paste("Histogram of the", g.netname,
-" Network"))
-legend("topright", c("Guess", "Poisson", "Least-Squares Fit",
-"Max Log-Likelihood"), lty=c(1,2,3,4), col=c("#40B0A6",
-"#006CD1", "#E66100", "#D35FB7"))
-g.mean <- mean(g.vec)
-g.seq <- 0:max(g.vec) # x-axis
-################# Guessing Alpha #################
-alpha.guess <- 1.5
-lines(g.seq, g.seq^(-alpha.guess), col="#40B0A6", lty=1, lwd=3)
-################# Poisson #################
-g.pois <- dpois(g.seq, g.mean, log=F)
-lines(g.seq, g.pois, col="#006CD1", lty=2)
-################# Linear model: Least-Squares Fit #################
-g.breaks <- g.hist$breaks[-c(1)] # remove 0
-g.probs <- g.hist$density[-1] # make lengths match
-# Need to clean up probabilities that are 0
-nz.probs.mask <- g.probs!=0
-g.breaks.clean <- g.breaks[nz.probs.mask]
-g.probs.clean <- g.probs[nz.probs.mask]
-#plot(log(g.breaks.clean), log(g.probs.clean))
-g.fit <- lm(log(g.probs.clean)~log(g.breaks.clean))
-summary(g.fit)
-alpha.LM <- coef(g.fit)[2]
-lines(g.seq, g.seq^(-alpha.LM), col="#E66100", lty=3)
-################# Max-Log-Likelihood #################
-n <- length(g.breaks.clean)
-kmin <- g.breaks.clean[1]
-alpha.ML <- 1 + n/sum(log(g.breaks.clean/kmin))
-alpha.ML
-lines(g.seq, g.seq^(-alpha.ML), col="#D35FB7", lty=4)
-# Homework 4 for the University of Tulsa' s CS-7863 Network Theory Course
-# Degree Distribution
-# Professor: Dr. McKinney, Spring 2022
-# Noah Schrick - 1492657
-library(igraph)
-library(igraphdata)
-data(yeast)
-g <- yeast
-g.netname <- "Yeast"
-################# Set up Work #################
-g.vec <- degree(g)
-g.hist <- hist(g.vec, freq=FALSE, main=paste("Histogram of the", g.netname,
-" Network"))
-legend("topright", c("Guess", "Poisson", "Least-Squares Fit",
-"Max Log-Likelihood"), lty=c(1,2,3,4), col=c("#40B0A6",
-"#006CD1", "#E66100", "#D35FB7"))
-g.mean <- mean(g.vec)
-g.seq <- 0:max(g.vec) # x-axis
-################# Guessing Alpha #################
-alpha.guess <- 1.5
-lines(g.seq, g.seq^(-alpha.guess), col="#40B0A6", lty=1, lwd=3)
-################# Poisson #################
-g.pois <- dpois(g.seq, g.mean, log=F)
-lines(g.seq, g.pois, col="#006CD1", lty=2, lwd=3)
-################# Linear model: Least-Squares Fit #################
-g.breaks <- g.hist$breaks[-c(1)] # remove 0
-g.probs <- g.hist$density[-1] # make lengths match
-# Need to clean up probabilities that are 0
-nz.probs.mask <- g.probs!=0
-g.breaks.clean <- g.breaks[nz.probs.mask]
-g.probs.clean <- g.probs[nz.probs.mask]
-#plot(log(g.breaks.clean), log(g.probs.clean))
-g.fit <- lm(log(g.probs.clean)~log(g.breaks.clean))
-summary(g.fit)
-alpha.LM <- coef(g.fit)[2]
-lines(g.seq, g.seq^(-alpha.LM), col="#E66100", lty=3, lwd=3)
-################# Max-Log-Likelihood #################
-n <- length(g.breaks.clean)
-kmin <- g.breaks.clean[1]
-alpha.ML <- 1 + n/sum(log(g.breaks.clean/kmin))
-alpha.ML
-lines(g.seq, g.seq^(-alpha.ML), col="#D35FB7", lty=4, lwd=3)
-plot(yeast)
-hist(yeast)
-hist(g.vec)
-g.pois
-g.mean
-alpha.LM
-alpha.ML
-degree(g)
-sort(degree(g))
-sort(degree(g),decreasing=FALSE)
-sort(degree(g),decreasing=F)
-sort(degree(g),decreasing=false)
-sort(degree(g), decreasing = TRUE)
-head(sort(degree(g), decreasing = TRUE))
-stddev(degree(g))
-sd(degree(g))
-tail(sort(degree(g), decreasing = TRUE))
-plot(log(g.breaks.clean), log(g.probs.clean))
-# Homework 4 for the University of Tulsa' s CS-7863 Network Theory Course
-# Degree Distribution
-# Professor: Dr. McKinney, Spring 2022
-# Noah Schrick - 1492657
-library(igraph)
-library(igraphdata)
-data(yeast)
-g <- yeast
-g.netname <- "Yeast"
-################# Set up Work #################
-g.vec <- degree(g)
-g.hist <- hist(g.vec, freq=FALSE, main=paste("Histogram of the", g.netname,
-" Network"))
-legend("topright", c("Guess", "Poisson", "Least-Squares Fit",
-"Max Log-Likelihood"), lty=c(1,2,3,4), col=c("#40B0A6",
-"#006CD1", "#E66100", "#D35FB7"))
-g.mean <- mean(g.vec)
-g.seq <- 0:max(g.vec) # x-axis
-################# Guessing Alpha #################
-alpha.guess <- 1.5
-lines(g.seq, g.seq^(-alpha.guess), col="#40B0A6", lty=1, lwd=3)
-################# Poisson #################
-g.pois <- dpois(g.seq, g.mean, log=F)
-lines(g.seq, g.pois, col="#006CD1", lty=2, lwd=3)
-################# Linear model: Least-Squares Fit #################
-g.breaks <- g.hist$breaks[-c(1)] # remove 0
-g.probs <- g.hist$density[-1] # make lengths match
-# Need to clean up probabilities that are 0
-nz.probs.mask <- g.probs!=0
-g.breaks.clean <- g.breaks[nz.probs.mask]
-g.probs.clean <- g.probs[nz.probs.mask]
-plot(log(g.breaks.clean), log(g.probs.clean))
-g.fit <- lm(log(g.probs.clean)~log(g.breaks.clean))
-summary(g.fit)
-alpha.LM <- coef(g.fit)[2]
-lines(g.seq, g.seq^(-alpha.LM), col="#E66100", lty=3, lwd=3)
-################# Max-Log-Likelihood #################
-n <- length(g.breaks.clean)
-kmin <- g.breaks.clean[1]
-alpha.ML <- 1 + n/sum(log(g.breaks.clean/kmin))
-alpha.ML
-lines(g.seq, g.seq^(-alpha.ML), col="#D35FB7", lty=4, lwd=3)
-plot(log(g.breaks.clean), log(g.probs.clean))
-g.breaks.clean <- g.breaks[nz.probs.mask]
-g.probs.clean <- g.probs[nz.probs.mask]
-plot(log(g.breaks.clean), log(g.probs.clean))
-graphvizCapabilities()$layoutTypes
-library(igraph)
-library(sna)
-library(Rgraphviz) # Reading graphviz' "dot" files
-graphvizCapabilities()$layoutTypes
-car.adj <- agread("./CG_Files/Network_1/DOTFILE.dot", layoutType="dot",layout=FALSE) # Large: ~1.9G
-################# Read in the previously generated networks #################
-# If sourcing:
-#setwd(getSrcDirectory()[1])
-# If running:
-setwd(dirname(rstudioapi::getActiveDocumentContext()$path))
-car.adj <- agread("./CG_Files/Network_1/DOTFILE.dot", layoutType="dot",layout=FALSE) # Large: ~1.9G
-plot(car.adj)
+V(hipaa.tc)$color <- ifelse(hipaa.tc.l_clusters>0, "green", "yellow")
+plot(hipaa.tc, main=paste(hipaa.tc.netname, "Laplace Spectral Clustering"), vertex.label=NA)
+pci.tc.eigs <- Re(eigen(pci.tc.Lap)$vectors[,pci.tc.n-1])
+pci.tc.eig_val <- eigen(pci.tc.Lap)$values[pci.tc.n-1]
+names(pci.tc.eigs) <- names(V(pci.tc))
+pci.tc.l_clusters <- ifelse(pci.tc.eigs>0,1,-1)
+tc_clusters[[3,1]] <- pci.tc.l_clusters
+V(pci.tc)$color <- ifelse(pci.tc.l_clusters>0, "green", "yellow")
+plot(pci.tc, main=paste(pci.tc.netname, "Laplace Spectral Clustering"), vertex.label=NA)
+### Clemente and Grassi
+tc_clusters[[1,2]] <- ClustBCG(as.matrix(car.tc.adj), "directed")$totalCC
+tc_clusters[[2,2]] <- ClustBCG(as.matrix(hipaa.tc.adj), "directed")$totalCC
+tc_clusters[[3,2]] <- ClustBCG(as.matrix(pci.tc.adj), "directed")$totalCC
+######################### Dominant Tree Centralities #########################
+dtree_centralities <- matrix(list(), nrow=3, ncol=5)
+rownames(dtree_centralities) <- c(car.netname, hipaa.netname, pci.netname)
+colnames(dtree_centralities) <- c("Degree", "Katz", "Page Rank", "K-path", "Betweenness")
+# Get basic network attributes
+car.dtree.adj <- get.adjacency(car.dtree)
+car.dtree.deg <- rowSums(as.matrix(car.dtree.adj)) # degree
+car.dtree.n <- length(V(car.dtree))
+hipaa.dtree.adj <- get.adjacency(hipaa.dtree)
+hipaa.dtree.deg <- rowSums(as.matrix(hipaa.dtree.adj)) # degree
+hipaa.dtree.n <- length(V(hipaa.dtree))
+pci.dtree.adj <- get.adjacency(pci.dtree)
+pci.dtree.deg <- rowSums(as.matrix(pci.dtree.adj)) # degree
+pci.dtree.n <- length(V(pci.dtree))
+### Degree
+dtree_centralities[[1,1]] <- car.dtree.deg %>% sort(decreasing = T)
+dtree_centralities[[2,1]] <- hipaa.dtree.deg %>% sort(decreasing = T)
+dtree_centralities[[3,1]] <- pci.dtree.deg %>% sort(decreasing = T)
+#### Katz
+car.dtree.katz <- katz.cent(car.dtree)
+dtree_centralities[[1,2]] <- car.dtree.katz[rowSums(apply(car.dtree.katz,2,is.nan))==0,] %>% sort(decreasing = T)
+dtree_centralities[[2,2]] <- katz.cent(hipaa.dtree) %>% sort(decreasing = T)
+dtree_centralities[[3,2]] <- katz.cent(pci.dtree) %>% sort(decreasing = T)
+### Page Rank
+dtree_centralities[[1,3]] <- page.rank(car.dtree)$vector %>% sort(decreasing = T)
+dtree_centralities[[2,3]] <- page.rank(hipaa.dtree)$vector %>% sort(decreasing = T)
+dtree_centralities[[3,3]] <- page.rank(pci.dtree)$vector %>% sort(decreasing = T)
+### K-path
+dtree_centralities[[1,4]] <- geokpath(car.dtree, V(car.dtree), "out") %>% sort(decreasing = T)
+dtree_centralities[[2,4]] <- geokpath(hipaa.dtree, V(hipaa.dtree), "out") %>% sort(decreasing = T)
+dtree_centralities[[3,4]] <- geokpath(pci.dtree, V(pci.dtree), "out") %>% sort(decreasing = T)
+### Betweenness
+dtree_centralities[[1,5]] <- betweenness(car.dtree, TRUE) %>% sort(decreasing = T)
+dtree_centralities[[2,5]] <- betweenness(hipaa.dtree, TRUE) %>% sort(decreasing = T)
+dtree_centralities[[3,5]] <- betweenness(pci.dtree, TRUE) %>% sort(decreasing = T)
+########################## Dominant Tree Clustering ##########################
+source("self_newman_mod.R")
+dtree_clusters <- matrix(list(), nrow=3, ncol=2)
+rownames(dtree_centralities) <- c(car.netname, hipaa.netname, pci.netname)
+colnames(dtree_centralities) <- c("Laplace", "CG")
+### Laplacian
+car.dtree.Lap <- diag(car.dtree.deg) - car.dtree.adj # L = D-A
+hipaa.dtree.Lap <- diag(hipaa.dtree.deg) - hipaa.dtree.adj
+pci.dtree.Lap <- diag(pci.dtree.deg) - pci.dtree.adj
+# get eigvals and vecs
+car.dtree.eigs <- Re(eigen(car.dtree.Lap)$vectors[,car.dtree.n-1])
+car.dtree.eig_val <- eigen(car.dtree.Lap)$values[car.dtree.n-1]
+names(car.dtree.eigs) <- names(V(car.dtree))
+car.dtree.l_clusters <- ifelse(car.dtree.eigs>0,1,-1)
+dtree_clusters[[1,1]] <- car.dtree.l_clusters
+V(car.dtree)$color <- ifelse(car.dtree.l_clusters>0, "green", "yellow")
+plot(car.dtree, main=paste(car.dtree.netname, "Laplace Spectral Clustering"), vertex.label=NA)
+hipaa.dtree.eigs <- Re(eigen(hipaa.dtree.Lap)$vectors[,hipaa.dtree.n-1])
+hipaa.dtree.eig_val <- eigen(hipaa.dtree.Lap)$values[hipaa.dtree.n-1]
+names(hipaa.dtree.eigs) <- names(V(hipaa.dtree))
+hipaa.dtree.l_clusters <- ifelse(hipaa.dtree.eigs>0,1,-1)
+dtree_clusters[[2,1]] <- hipaa.dtree.l_clusters
+V(hipaa.dtree)$color <- ifelse(hipaa.dtree.l_clusters>0, "green", "yellow")
+plot(hipaa.dtree, main=paste(hipaa.dtree.netname, "Laplace Spectral Clustering"), vertex.label=NA)
+pci.dtree.eigs <- Re(eigen(pci.dtree.Lap)$vectors[,pci.dtree.n-1])
+pci.dtree.eig_val <- eigen(pci.dtree.Lap)$values[pci.dtree.n-1]
+names(pci.dtree.eigs) <- names(V(pci.dtree))
+pci.dtree.l_clusters <- ifelse(pci.dtree.eigs>0,1,-1)
+dtree_clusters[[3,1]] <- pci.dtree.l_clusters
+V(pci.dtree)$color <- ifelse(pci.dtree.l_clusters>0, "green", "yellow")
+plot(pci.dtree, main=paste(pci.dtree.netname, "Laplace Spectral Clustering"), vertex.label=NA)
+### Clemente and Grassi
+dtree_clusters[[1,2]] <- ClustBCG(as.matrix(car.dtree.adj), "directed")$totalCC
+dtree_clusters[[2,2]] <- ClustBCG(as.matrix(hipaa.dtree.adj), "directed")$totalCC
+dtree_clusters[[3,2]] <- ClustBCG(as.matrix(pci.dtree.adj), "directed")$totalCC
+############################# Write Final Results #############################
+write.table(base_centralities, file='results.csv')
+write.table(tc_centralities, file='results.csv')
+write.table(dtree_centralities, file='results.csv')
+### Degree:
+head(base_centralities[[1,1]], 15) #Car
+head(tc_centralities[[1,1]],15)
+head(dtree_centralities[[1,1]],15)
+head(base_centralities[[2,1]], 15) #HIPAA
+head(tc_centralities[[2,1]],15)
+head(dtree_centralities[[2,1]],15)
+head(base_centralities[[3,1]], 15) #PCI
+head(tc_centralities[[3,1]],15)
+head(dtree_centralities[[3,1]],15)
+### Katz:
+head(base_centralities[[1,2]], 15) #Car
+head(tc_centralities[[1,2]],15)
+head(dtree_centralities[[1,2]],15)
+head(base_centralities[[2,2]], 15) #HIPAA
+head(tc_centralities[[2,2]],15)
+head(dtree_centralities[[2,2]],15)
+head(base_centralities[[3,2]], 15) #PCI
+head(tc_centralities[[3,2]],15)
+head(dtree_centralities[[3,2]],15)
+### Page Rank:
+head(base_centralities[[1,3]], 15) #Car
+head(tc_centralities[[1,3]],15)
+head(dtree_centralities[[1,3]],15)
+head(base_centralities[[2,3]], 15) #HIPAA
+head(tc_centralities[[2,3]],15)
+head(dtree_centralities[[2,3]],15)
+head(base_centralities[[3,3]], 15) #PCI
+head(tc_centralities[[3,3]],15)
+head(dtree_centralities[[3,3]],15)
+### K-Path:
+head(base_centralities[[1,4]], 15) #Car
+head(tc_centralities[[1,4]],15)
+head(dtree_centralities[[1,4]],15)
+head(base_centralities[[2,4]], 15) #HIPAA
+head(tc_centralities[[2,4]],15)
+head(dtree_centralities[[2,4]],15)
+head(base_centralities[[3,4]], 15) #PCI
+head(tc_centralities[[3,4]],15)
+head(dtree_centralities[[3,4]],15)
+### Betweenness:
+head(base_centralities[[1,5]], 15) #Car
+head(tc_centralities[[1,5]],15)
+head(dtree_centralities[[1,5]],15)
+head(base_centralities[[2,5]], 15) #HIPAA
+head(tc_centralities[[2,5]],15)
+head(dtree_centralities[[2,5]],15)
+head(base_centralities[[3,5]], 15) #PCI
+head(tc_centralities[[3,5]],15)
+head(dtree_centralities[[3,5]],15)
+### Laplacian:
+head(base_clusters[[1,1]], 15) #Car
+head(tc_centralities[[1,1]],15)
+head(dtree_centralities[[1,1]],15)
+head(base_clusters[[2,1]], 15) #HIPAA
+head(tc_centralities[[2,1]],15)
+head(dtree_centralities[[2,1]],15)
+head(base_clusters[[3,1]], 15) #PCI
+head(tc_centralities[[3,1]],15)
+head(dtree_centralities[[3,1]],15)
+### CG:
+head(base_clusters[[1,2]], 15) #Car
+head(tc_centralities[[1,2]],15)
+head(dtree_centralities[[1,2]],15)
+head(base_clusters[[2,2]], 15) #HIPAA
+head(tc_centralities[[2,2]],15)
+head(dtree_centralities[[2,2]],15)
+head(base_clusters[[3,2]], 15) #PCI
+head(tc_centralities[[3,2]],15)
+head(dtree_centralities[[3,2]],15)
+head(base_centralities[[2,2]], 15) #HIPAA
+### Page Rank:
+head(base_centralities[[1,3]], 15) #Car
+### Page Rank:
+as.matrix(head(base_centralities[[1,3]], 15) #Car)
+)
+### Page Rank:
+as.matrix(head(base_centralities[[1,3]], 15))[2]
+### Page Rank:
+as.matrix(head(base_centralities[[1,3]], 15))
+### Page Rank:
+as.matrix(head(base_centralities[[1,3]], 15))[,1]
+### Page Rank:
+as.matrix(head(base_centralities[[1,3]], 15))[,2]
+### Page Rank:
+as.matrix(head(base_centralities[[1,3]], 15))[1,]
+### Page Rank:
+as.matrix(head(base_centralities[[1,3]], 15))[2,]
+### Page Rank:
+as.matrix(head(base_centralities[[1,3]], 15))
+### Page Rank:
+as.matrix(head(tc_centralities[[1,3]], 15))
+### Page Rank:
+as.matrix(head(base_centralities[[2,3]], 15))
+### Page Rank:
+as.matrix(head(tc_centralities[[2,3]], 15))
+### Page Rank:
+as.matrix(head(dtree_centralities[[2,3]], 15))
+### Page Rank:
+as.matrix(head(dtree_centralities[[3,3]], 15))
+### Page Rank:
+as.matrix(head(tc_centralities[[3,3]], 15))
+### Page Rank:
+as.matrix(head(base_centralities[[3,3]], 15))
+head(tc_centralities[[1,4]],15)
+### Page Rank:
+as.matrix(as.matrix(head(base_centralities[[3,3]], 15)))
+as.matrix(head(tc_centralities[[1,4]],15))
+as.matrix(head(base_centralities[[2,4]],15))
+as.matrix(head(tc_centralities[[2,4]],15))
+as.matrix(head(dtree_centralities[[2,4]],15))
+as.matrix(head(base_centralities[[3,4]],15))
+as.matrix(head(tc_centralities[[3,4]],15))
+as.matrix(head(dtree_centralities[[3,4]],15))
+as.matrix(head(base_centralities[[1,5]], 15))
+as.matrix(head(tc_centralities[[1,5]], 15))
+as.matrix(head(dtree_centralities[[1,5]], 15))
+as.matrix(head(dtree_centralities[[2,5]], 15))
+as.matrix(head(tc_centralities[[2,5]], 15))
+as.matrix(head(base_centralities[[2,5]], 15))
+as.matrix(head(base_centralities[[3,5]], 15))
+as.matrix(head(base_centralities[[2,5]], 15))
+as.matrix(head(tc_centralities[[3,5]], 15))
+as.matrix(head(dtree_centralities[[3,5]], 15))
+edge_connectivity(car)
+edge_connectivity(hipaa)
+edge_connectivity(pci)
+edge_connectivity(car, "0", "2490")
+Vcount(car)
+ecount(car)
+vcount(car)
+ecount(hipaa)
+vcount(hipaa)
+vcount(pci)
+ecount(pci\)
+ecount(pci)
+mean(base_centralities[[1,1]])
+mean(base_centralities[[1,1]])/vcount(car)
+100*mean(base_centralities[[1,1]])/vcount(car)
+100*mean(base_centralities[[2,1]])/vcount(hipaa)
+100*mean(base_centralities[[3,1]])/vcount(pci)
+katzcent(car,,0.9)
+katzcent(car,,0.1)
+as.matrix(head(katzcent(car,,0.1) %>% sort(decreasing=T), 15))
+as.matrix(head(katzcent(car.tc,,0.1) %>% sort(decreasing=T), 15))
+as.matrix(head(katzcent(car.dtree,,0.1) %>% sort(decreasing=T), 15))
+as.matrix(head(katzcent(car.dtree,,0.1) %>% sort(decreasing=T), 45))
+as.matrix(head(katzcent(hipaa,,0.1) %>% sort(decreasing=T), 15))
+as.matrix(head(Re(katzcent(hipaa,,0.1)) %>% sort(decreasing=T), 15))
+katzcent(hipaa,,0.1)
+as.matrix(head(katzcent(pci,,0.1) %>% sort(decreasing=T), 15))
+katzcent(pci,,0.1)
+katz.cent(pci)
+katz.cent(pci) %>% sort(decreasing = %)
+katz.cent(pci) %>% sort(decreasing = T)
+head(base_centralities[[2,2]], 15) #HIPAA
+base_centralities[[2,2]]
+order(base_centralities[[2,2]],decreasing=T)
+order(tc_centralities[[2,2]],decreasing=T)
+tc_centralities[[2,2]]
+katz.cent(pci)
+order(katz.cent(pci))
+sort(katz.cent(pci))[60]
+sort(katz.cent(pci))[47]
+sort(katz.cent(pci))[61]
+sort(katz.cent(pci))[56]
+sort(katz.cent(pci))[54]
+sort(katz.cent(pci))[58]
+sort(katz.cent(pci))[53]
+sort(katz.cent(pci))[57]
+sort(katz.cent(pci))[45]
+sort(katz.cent(pci))[44]
+sort(katz.cent(pci))[32]
+sort(katz.cent(pci))
+order(katz.pci, decreasing=T)
+order(katz.cent(pci), decreasing=T)
+head(order(katz.cent(pci), decreasing=T), 15)
+katz.cent(pci)[1]
+katz.cent(pci)[2]
+katz.cent(pci)[5]
+katz.cent(pci)[12]
+katz.cent(pci)[23]
+katz.cent(pci)[4]
+katz.cent(pci)[11]
+katz.cent(pci)[22]
+katz.cent(pci)[36]
+katz.cent(pci)[3]
+katz.cent(pci)[9]
+katz.cent(pci)[20]
+katz.cent(pci)[34]
+katz.cent(pci)[39]
+katz.cent(pci)[26]
+order(katz.cent(pci.tc), decreasing=T)
+order(katz.cent(pci.tc), decreasing=T) %>% sort()
+order(katz.cent(pci.tc), decreasing=T) %>% sort(katz.cent(pci.tc),decreasing=T)[]
+order(katz.cent(pci.tc), decreasing=T) %>%katz.cent[]
+order(katz.cent(pci.tc), decreasing=T) %>%katz.cent(pci.tc)[]
+order(katz.cent(pci.tc), decreasing=T)
+katz.cent(pci.tc)[0]
+katz.cent(pci.tc)[1]
+katz.cent(pci.tc)[2]
+katz.cent(pci.tc)[6]
+katz.cent(pci.tc)[5]
+katz.cent(pci.tc)[12]
+katz.cent(pci.tc)[23]
+katz.cent(pci.tc)[3]
+katz.cent(pci.tc)[4]
+katz.cent(pci.tc)[9]
+katz.cent(pci.tc)[20]
+katz.cent(pci.tc)[34]
+katz.cent(pci.tc)[11]
+katz.cent(pci.tc)[22]
+katz.cent(pci.tc)[36]
+katz.cent(pci.tc)[8]
+katz.cent(pci.tc)[39]
+order(katz.cent(pci.dtree), decreasing=T
+)
+order(katz.cent(pci.dtree), decreasing=T)
+order(katz.cent(pci.tc), decreasing=T)
+order(katz.cent(pci.dtree), decreasing=T)
+katz.cent(pci.dtree)
+katzcent(pci.dtree)
+katzcent(pci.dtree,,0.1)
+katzcent(pci.dtree,,0.9)
+katzcent(pci.dtree,,0.2)
+katzcent(pci.dtree,,0.1)
+katz.cent(pci.dtree)
+katz.cent.adj
+pci.dtree.adj
+eigen(pci.dtree.adj)$values[1]
+katz.cent(pci.dtree, 0.9)
+katz.cent(pci.tc, 0.9)
+katz.cent(pci.tc, 0.1)
+katz.cent(pci.dtree, 0.1)
+sort(katz.cent(pci.dtree, 0.1), decreasing=T)
+sort(katz.cent(pci.dtree, 0.1), decreasing=T, index.return=T)
+sort(katz.cent(pci.dtree, 0.1), decreasing=T, index.return=T)$x
+sort(katz.cent(pci.dtree, 0.1), decreasing=T, index.return=T)$ix
+sort(katz.cent(pci.dtree, 0.1), decreasing=T, index.return=TRUE)
+sort(katz.cent(pci.dtree, 0.1), decreasing=T, index.return=TRUE)$x
+sort.index(katz.cent(pci.dtree,0.1),decreasing=T)
+tmp <- katz.cent(pci.dtree,0.1)
+sort(tmp, decreasing=T, index.return=T)
+tmp$x
+tmp2 <- sort(tmp, decreasing=T, index.return=T)
+tmp2
+tmp2$x
+tmp2$ix
+tmp2[1]
+tmp2 <- sort(tmp, decreasing=T, index.return=TRUE)
+tmp2$ix
+tmp2
+tmp2 <- sort(tmp, index.return=TRUE, decreasing=T)
+tmp21
+tmp2
+tmp2$ix
+tmp2 <- sort(tmp, index.return=TRUE)
+tmp2
+tmp2$ix
+tmp2$x
+tmp <- matrix(list(), nrow=vcount(car), ncol=2)
+tmp[,1] <- car.katz %>% sort(,decreasing=T)
+tmp[,1] <- car.katz %>% sort(decreasing=T)
+tmp[1] <- car.katz %>% sort(decreasing=T)
+tmp[[,1]] <- car.katz %>% sort(decreasing=T)
+nodes <- car.katz %>% sort(decreasing=T)
+car.katz
+car.katz(car,.9)
+car.katz<- katz.cent(car,.9)
+car.katz[2488]
+car.katz[1]
+car.katz[5]
+car.katz
+is.na(eigen(car.adj)$values[1])
+eigen(car.adj)$values[1]
+eigen(car.adj)$values[1] == 0
+############################# Base Centralities #############################
+source("centralities.R")
+############################# Base Centralities #############################
+source("centralities.R")
+katz.cent(car)
+############################# Base Centralities #############################
+source("centralities.R")
+katz.cent(car)
+#### Katz
+car.katz <- katz.cent(car)
+nodes <- car.katz %>% sort(decreasing=T)
+vals <- car.katz %>% order(decreasing=T)
+head(nodes,15)
+as.matrix(head(nodes,15))
+as.matrix(head(nodes,15), head(vals, 15))
+as.matrix(head(vals,15))
+as.matrix(head(vals,15)-1)
+as.data.frame(head(vals,15)-1, head(nodes,15))
+as.data.frame(head(vals,15), head(nodes,15))
+nodes <- car.katz %>% sort(decreasing=T)
+nodes <- car.katz %>% order(decreasing=T)
+vals <- car.katz %>% sort(decreasing=T)
+head(nodes,15)
+head(vals,15)
+as.data.frame(head(nodes,15), head(vals, 15))
+nodes <- car.katz %>% order(decreasing=T)
+nodes <- head(nodes, 15)
+nodes
+nodes <- car.katz %>% order(decreasing=T)
+nodes <- head(nodes, 15)-1
+nodes
+vals <- head(vals, 15)
+as.data.frame(nodes, vals)
+as.matrix(nodes, vals)
+print(cbind(nodes, vals))
+print(cbind(nodes, vals))$nodes
+tmp <- (cbind(nodes, vals))
+tmp
+prmatrix(tmp)
+nodes
+as.matrix(nodes)
+as.matrix(nodes)[1]
+as.matrix(nodes)[,1]
+vals
+as.matrix(vals)
+#### Katz
+car.katz <- katz.cent(car)
+nodes <- car.katz %>% order(decreasing=T)
+nodes <- head(nodes, 15)-1
+vals <- car.katz %>% sort(decreasing=T)
+vals <- head(vals, 15)
+nodes
+vals
+as.matrix(vals)
+############################# Base Centralities #############################
+source("centralities.R")
+#### Katz
+car.katz <- katz.cent(car)
+nodes <- car.katz %>% order(decreasing=T)
+nodes <- head(nodes, 15)-1
+vals <- car.katz %>% sort(decreasing=T)
+vals <- head(vals, 15)
+nodes
+vals
+as.matrix(nodes)
+as.matrix(vals)
+hipaa.katz <- katz.cent(hipaa)
+nodes <- hipaa.katz %>% order(decreasing=T)
+nodes <- head(nodes, 15)-1
+vals <- hipaa.katz %>% sort(decreasing=T)
+vals <- head(vals, 15)
+as.matrix(nodes)
+as.matrix(vals)
+pci.katz <- katz.cent(pci)
+nodes <- pci.katz %>% order(decreasing=T)
+nodes <- head(nodes, 15)-1
+vals <- pci.katz %>% sort(decreasing=T)
+vals <- head(vals, 15)
+as.matrix(nodes)
+as.matrix(vals)
+car.tc.katz <- katz.cent(car.tc)
+nodes <- car.tc.katz %>% order(decreasing=T)
+nodes <- head(nodes, 15)-1
+vals <- car.tc.katz %>% sort(decreasing=T)
+vals <- head(vals, 15)
+as.matrix(nodes)
+as.matrix(vals)
+hipaa.tc.katz <- katz.cent(hipaa.tc)
+plot(pci.dtree)
+V(pci.dtree)$color <- "yellow"
+plot(pci.dtree)
+hipaa.tc.katz <- katz.cent(hipaa.tc)
+nodes <- hipaa.tc.katz %>% order(decreasing=T)
+nodes <- head(nodes, 15)-1
+vals <- hipaa.tc.katz %>% sort(decreasing=T)
+vals <- head(vals, 15)
+as.matrix(nodes)
+as.matrix(vals)
+pci.tc.katz <- katz.cent(pci.tc)
+nodes <- pci.tc.katz %>% order(decreasing=T)
+nodes <- head(nodes, 15)-1
+vals <- pci.tc.katz %>% sort(decreasing=T)
+vals <- head(vals, 15)
+as.matrix(nodes)
+as.matrix(vals)
+car.dtree.katz <- katz.cent(car.dtree)
+nodes <- car.dtree.katz %>% order(decreasing=T)
+nodes <- head(nodes, 15)-1
+vals <- car.dtree.katz %>% sort(decreasing=T)
+vals <- head(vals, 15)
+as.matrix(nodes)
+as.matrix(vals)
+hipaa.dtree.katz <- katz.cent(hipaa.dtree)
+nodes <- hipaa.dtree.katz %>% order(decreasing=T)
+nodes <- head(nodes, 15)-1
+vals <- hipaa.dtree.katz %>% sort(decreasing=T)
+vals <- head(vals, 15)
+as.matrix(nodes)
+as.matrix(vals)
+pci.dtree.katz <- katz.cent(pci.dtree)
+nodes <- pci.dtree.katz %>% order(decreasing=T)
+nodes <- head(nodes, 15)-1
+nodes <- head(nodes, 15)-1
+vals <- pci.dtree.katz %>% sort(decreasing=T)
+vals <- head(vals, 15)
+as.matrix(nodes)
+nodes <- pci.dtree.katz %>% order(decreasing=T)
+pci.dtree.katz <- katz.cent(pci.dtree)
+nodes <- pci.dtree.katz %>% order(decreasing=T)
+nodes <- head(nodes, 15)-1
+vals <- pci.dtree.katz %>% sort(decreasing=T)
+vals <- head(vals, 15)
+as.matrix(nodes)
+as.matrix(vals)
+plot(pci)
+V(pci)$color <- "yellow"
+plot(pci)
+plot(pci.tc)
+V(pci.tc)$color <- "yellow"
+plot(pci.tc)
+plot(dtree_clusters[[1,2]])
+base_clusters[[1,2]]
+which(base_clusters[[1,2]] > 1)
+which(base_clusters[[1,2]] > 0)
+which(base_clusters[[2,2]] > 0)
+which(base_clusters[[3,2]] > 0)
+plot(base_clusters[[3,2]]
+)
+pci.dtree.deg
diff --git a/Presentation/Schrick-Noah_CS-7863_Final-Presentation.odp b/Presentation/Schrick-Noah_CS-7863_Final-Presentation.odp
new file mode 100644
index 0000000..8f954d3
Binary files /dev/null and b/Presentation/Schrick-Noah_CS-7863_Final-Presentation.odp differ
diff --git a/Presentation/Schrick-Noah_CS-7863_Final-Presentation.pdf b/Presentation/Schrick-Noah_CS-7863_Final-Presentation.pdf
new file mode 100644
index 0000000..7c447f0
Binary files /dev/null and b/Presentation/Schrick-Noah_CS-7863_Final-Presentation.pdf differ
diff --git a/Presentation/Schrick-Noah_CS-7863_Final-Presentation.ppt b/Presentation/Schrick-Noah_CS-7863_Final-Presentation.ppt
new file mode 100644
index 0000000..54a7456
Binary files /dev/null and b/Presentation/Schrick-Noah_CS-7863_Final-Presentation.ppt differ
diff --git a/Presentation/images/APC.png b/Presentation/images/APC.png
new file mode 100644
index 0000000..11a59f1
Binary files /dev/null and b/Presentation/images/APC.png differ
diff --git a/Presentation/images/Betweenness.png b/Presentation/images/Betweenness.png
new file mode 100644
index 0000000..ba418b6
Binary files /dev/null and b/Presentation/images/Betweenness.png differ
diff --git a/Presentation/images/Car-Betwn.png b/Presentation/images/Car-Betwn.png
new file mode 100644
index 0000000..c4a01e9
Binary files /dev/null and b/Presentation/images/Car-Betwn.png differ
diff --git a/Presentation/images/Car-Deg.png b/Presentation/images/Car-Deg.png
new file mode 100644
index 0000000..e2296af
Binary files /dev/null and b/Presentation/images/Car-Deg.png differ
diff --git a/Presentation/images/KPE.png b/Presentation/images/KPE.png
new file mode 100644
index 0000000..0e4a4f7
Binary files /dev/null and b/Presentation/images/KPE.png differ
diff --git a/Presentation/images/Katz1.png b/Presentation/images/Katz1.png
new file mode 100644
index 0000000..86b18d2
Binary files /dev/null and b/Presentation/images/Katz1.png differ
diff --git a/Presentation/images/Katz2.png b/Presentation/images/Katz2.png
new file mode 100644
index 0000000..32a5436
Binary files /dev/null and b/Presentation/images/Katz2.png differ
diff --git a/Presentation/images/Network_Table.png b/Presentation/images/Network_Table.png
new file mode 100644
index 0000000..f8669d5
Binary files /dev/null and b/Presentation/images/Network_Table.png differ
diff --git a/Presentation/images/PR.png b/Presentation/images/PR.png
new file mode 100644
index 0000000..226955d
Binary files /dev/null and b/Presentation/images/PR.png differ
diff --git a/Presentation/images/PageRank.png b/Presentation/images/PageRank.png
new file mode 100644
index 0000000..c5dee1d
Binary files /dev/null and b/Presentation/images/PageRank.png differ
diff --git a/README.md b/README.md
index e69de29..86f45cd 100644
--- a/README.md
+++ b/README.md
@@ -0,0 +1,9 @@
+Compliance Graph Analysis Techniques using Network Theory Approaches.
+
+CG_Files/ contains the network models, exploit files, dotfiles, and edge text files for each network. In addition, a manual import of the networks is given in an R script due to difficulties of variable importing of networks.
+
+Report/ contains various tex files used for the generation of the report.
+
+Presentation/ contains the various files used for the creation of the presentation.
+
+The main root folder contains various R files used as supporting functions for the main "Schrick-Noah_CG-Analysis.R" file.
diff --git a/Submission/Schrick-Noah_CS-7863_Final-Presentation.pdf b/Submission/Schrick-Noah_CS-7863_Final-Presentation.pdf
new file mode 100644
index 0000000..7c447f0
Binary files /dev/null and b/Submission/Schrick-Noah_CS-7863_Final-Presentation.pdf differ
diff --git a/Submission/Schrick-Noah_CS-7863_Final-Presentation.ppt b/Submission/Schrick-Noah_CS-7863_Final-Presentation.ppt
new file mode 100644
index 0000000..54a7456
Binary files /dev/null and b/Submission/Schrick-Noah_CS-7863_Final-Presentation.ppt differ
diff --git a/Submission/Schrick-Noah_CS-7863_Final-Report.pdf b/Submission/Schrick-Noah_CS-7863_Final-Report.pdf
new file mode 100644
index 0000000..4a929ca
Binary files /dev/null and b/Submission/Schrick-Noah_CS-7863_Final-Report.pdf differ