Ngram Difference Between Days Script

scripts/ngram Difference between Days X and Y.R

@@ -0,0 +1,97 @@
+# Load Libraries
+library("ggplot2")
+library("tidyverse")
+library("dplyr")
+# For the %>% operator, but you can
+# skip loading tidyverse 
+# and just use `|>` as 
+# a pipe operator.
+
+
+# Note: Other code is below to do an alternative method,
+#       but the uncommented method is superior.
+
+#load CSVs using code. 
+df1 <- read.csv("~/Documents/Stats/4ChanScraper/ngram Aug 31 2023 10:49:52.csv")
+df2 <- read.csv("~/Documents/Stats/4ChanScraper/ngram Aug 31 2023 12:08:28.csv")
+
+
+# Merge data frame, and take difference b/w day 1 and day 2
+# subtracting data frames from each other.
+# X.x = df1
+# X.y = df2
+df_merged <- merge(df1, df2, by="word", all=TRUE)
+# For ngram below
+df_merged$result <- df_merged$X.y - df_merged$X.x
+
+
+# Feel free to add more "non-words," or "noise" 
+# to this list as you see fit.
+df_difference_filter <- df_merged %>% 
+  filter(!word == "de"
+         & !word == "je"
+         & !word == "een"
+         & !word == "dat"
+         & !word == "en"
+         & !word == "eu"
+         & !word == "te"
+         & !word == "tu"
+         & !word == "niet"
+         & !word == "van"
+         & !word == "niet"
+         & !word == "ik"
+         & !word == "ze"
+         & !word == "om"
+         & !word == "met"
+         & !word == "uk"
+         & !word == "qt"
+         & !word == "wat"
+         & !word == "bb"
+         & !word == "op"
+         & !word == "ne"
+         & !word == "rh")
+
+# assign NA to Zero
+df_difference_filter$result[is.na(df_difference_filter$result)] <- 0
+
+
+# Get bottom 20 (negative) numbers
+df_bottom <- df_difference_filter %>% 
+  top_n(-20)
+
+# Get top 20 (positive) numbers
+df_top <- df_difference_filter %>% 
+  top_n(20)
+
+# Bind into new data frame
+df_merged2 <- rbind(df_top, df_bottom)
+
+
+# Colours results based on conditions.
+ngram_fill_bar <- case_when(
+  df_merged2$result < -22000   ~ "red4",
+  df_merged2$result <= -20000 ~ "red1",
+  df_merged2$result <= -19000 ~ "darkgreen",
+  df_merged2$result <= -18000 ~ "seagreen",
+  df_merged2$result <= 0 ~ "black",
+  df_merged2$result <= 20000 ~ "slategray4",
+  df_merged2$result <= 21000 ~ "slateblue",
+  df_merged2$result <= 23000 ~ "steelblue3",
+  df_merged2$result <= 25000 ~ "cyan4")
+
+
+# bar graph of difference between Day 2, and Day 1.
+df_merged2 %>% 
+  top_n(50) %>% 
+  mutate(word = reorder(word, result)) %>%
+  ggplot(aes(word, result)) +
+  theme(legend.position = "none", axis.title.y = element_blank()) +
+  geom_bar(stat = "identity", fill = ngram_fill_bar) +
+  labs(
+    title = "Difference of Word Pairs from Today - Yesterday",
+    x = "Words",
+    y = "Count",
+    caption = "Positive integers = More mentions today.
+     Negative integers = Less mentions today.") +
+  coord_flip() +
+  theme_dark(base_size = 13)