if (!require("ggplot2")) install.packages("ggplot2")
library(ggplot2)
if (!require("reshape2")) install.packages("reshape2")
library(reshape2)

data <- read.csv("Word_Dynamics2.csv")

keywords <- data$keyword

data_without_keywords <- t(data[,-1])

normalize <- function(x) {
  return((x - min(x)) / (max(x) - min(x)))
}

normalized_data <- apply(data_without_keywords, 2, normalize)

cor_matrix <- cor(normalized_data)
colnames(cor_matrix) <- keywords
rownames(cor_matrix) <- keywords

hc <- hclust(dist(cor_matrix))

dendrogram <- as.dendrogram(hc)

ordered_cor_matrix <- cor_matrix[order.dendrogram(dendrogram), order.dendrogram(dendrogram)]

melted_ordered_cor_matrix <- melt(ordered_cor_matrix)

p <- ggplot(data = melted_ordered_cor_matrix, aes(x=Var1, y=Var2, fill=value)) + 
  geom_tile() +
  scale_fill_gradient2(low = "blue", high = "red", mid = "white", midpoint = 0) +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 90, hjust = 1, size = 8),  
        axis.text.y = element_text(angle = 0, hjust = 1, size = 8)) + 
  labs(fill = "Correlation", x = "Keyword", y = "Keyword")
p
ggsave("Correlation.png", plot = p, width = 12, height = 12, dpi = 300)

install.packages("reshape2")
if (!requireNamespace("BiocManager", quietly = TRUE))
  install.packages("BiocManager")

BiocManager::install("ComplexHeatmap")
install.packages("circlize")
library(reshape2)
library(ComplexHeatmap)
library(circlize)

data <- read.csv("Word_Dynamics2.csv")

data_long <- melt(data, id.vars = "keyword")
data_long$value <- as.numeric(data_long$value) 

data_long <- ddply(data_long, .(keyword), transform, scale = scale(value))

data_wide <- dcast(data_long, keyword ~ variable, value.var = "scale")
colnames(data_wide)[-1] <- gsub("^X", "", colnames(data_wide)[-1]) 

rownames(data_wide) <- data_wide$keyword
data_matrix <- as.matrix(data_wide[,-1])

my_col_fun <- colorRamp2(c(-2, 0, 2), c("blue", "white", "red"))

pdf("heatmap_output.pdf", width = 10, height = 12)

ht <- Heatmap(data_matrix, 
              name = "scale", 
              col = my_col_fun, 
              cluster_rows = TRUE,
              show_row_names = TRUE, 
              row_names_side = "right", 
              cluster_columns = FALSE, 
              row_names_gp = gpar(fontsize = 10), 
              row_dend_width = unit(2, "cm"), 
              clustering_distance_rows = "euclidean", 
              clustering_method_rows = "complete", 
              row_names_max_width = unit(8, "cm") 
)

draw(ht, heatmap_legend_side = "bot", annotation_legend_side = "bot")

dev.off()

install.packages("ggplot2")
install.packages("reshape2")
install.packages("dendextend")
install.packages("gplots")
library(ggplot2)
library(reshape2)
library(plyr)
library(dendextend)
library(gplots)
library(stringr)
library(reshape2)

data <- read.csv("Word_Dynamics2.csv")

data_long <- melt(data, id.vars = "keyword")
data_long$value <- as.numeric(data_long$value) 
data_long <- ddply(data_long, .(keyword), transform, scale = scale(value))

data_long$variable <- str_replace(data_long$variable, "^X", "")

p <- ggplot(data_long, aes(x = variable, y = keyword, fill = scale)) + 
  geom_tile() + 
  scale_fill_gradient2(low = "blue", mid = "white", high = "red", midpoint = 0) +
  theme_minimal() +  
  xlab("Year") + 
  ylab("Keyword") +
  ggtitle("Row Scaled Heatmap of Keyword Dynamics") +
  theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust = 1),
        plot.background = element_rect(fill = "transparent"))  
p
ggsave("heatmap.png", plot = p, width = 9, height = 12, dpi = 300)


install.packages("ggplot2")
library(ggplot2)
data <- read.csv("Word_Dynamics2.csv")
keywords <- data$keyword
data_without_keywords <- data[,-1]
normalize <- function(x) {
  return((x - min(x)) / (max(x) - min(x)))
}
normalized_data <- as.data.frame(lapply(data_without_keywords, normalize))
transposed_normalized_data <- t(normalized_data)
cor_matrix <- cor(transposed_normalized_data, method = "pearson")
rownames(cor_matrix) <- colnames(cor_matrix) <- keywords
melted_cor_matrix <- melt(cor_matrix)
ggplot(data = melted_cor_matrix, aes(x=Var1, y=Var2, fill=value)) + 
  geom_tile() +
  scale_fill_gradient2(low = "blue", high = "red", mid = "white", midpoint = 0) +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1), 
        axis.text.y = element_text(angle = 45, hjust = 1)) +
  labs(fill = "Correlation", x = "Keyword", y = "Keyword")

install.packages("ggplot2")
install.packages("drc")
install.packages("minpack.lm")
install.packages("minpack.lm")
library(ggplot2)
library(dplyr)
library(broom)
library(minpack.lm)
getwd()
data <- read.csv("Logistic.csv")


logistic_fit <- nlsLM(Documents ~ SSlogis(Year, Asym, xmid, scal), data = data)
poly_fit2 <- lm(Documents ~ poly(Year, 2), data = data)

year_seq <- seq(min(data$Year), max(data$Year) + 10)
predictions <- expand.grid(Year = year_seq)


predictions <- predictions %>%
  mutate(
    Logistic = predict(logistic_fit, newdata = predictions),
    Poly2 = predict(poly_fit2, newdata = predictions, interval = "predict")[, 1]
  )

pred_intervals <- predict(poly_fit2, newdata = predictions, interval = "predict")
predictions$lwr <- pred_intervals[, "lwr"]
predictions$upr <- pred_intervals[, "upr"]
p<-ggplot() +
  geom_point(data = data, aes(x = Year, y = Documents)) +  
  geom_line(data = predictions, aes(x = Year, y = Logistic), color = "red") + 
  geom_ribbon(data = predictions, aes(x = Year, ymin = lwr, ymax = upr), fill = "red", alpha = 0.2) +  
  geom_line(data = predictions, aes(x = Year, y = Poly2), color = "blue") +  
  theme_minimal() +
  labs(title = "各模型拟合及预测区间")
p
ggsave("logistic.png", plot = p, width = 6, height = 6, dpi = 300)