From d6be2a2ce998f492ddf347a3f7e5fec18627815b Mon Sep 17 00:00:00 2001
From: Kuba <jakeic@st.amu.edu.pl>
Date: Tue, 21 Mar 2023 22:31:17 +0100
Subject: [PATCH] calculate frequency of bigrams in text

---
 analysis/bigrams_freq.sh | 4 ++++
 1 file changed, 4 insertions(+)
 create mode 100755 analysis/bigrams_freq.sh

diff --git a/analysis/bigrams_freq.sh b/analysis/bigrams_freq.sh
new file mode 100755
index 0000000..510e1ab
--- /dev/null
+++ b/analysis/bigrams_freq.sh
@@ -0,0 +1,4 @@
+#!/bin/bash
+
+# Convert text to lowercase and replace non-alphabetic characters with spaces, then use awk to print bigrams
+bzcat $1 | tr '[:upper:]' '[:lower:]' | tr -cs '[:alpha:]' ' ' | fold -w 1000 | sed -e 's/\b\s\+\b/ /g' | awk '{for(i=1;i<NF;i++)print $i,$(i+1)}' | sort | uniq -c > bigram_freq.txt
\ No newline at end of file