diff --git a/scripts/dataset/build_dataset.sh b/scripts/dataset/build_dataset.sh
new file mode 100755
index 0000000..1449fc4
--- /dev/null
+++ b/scripts/dataset/build_dataset.sh
@@ -0,0 +1,27 @@
+#!/bin/bash
+
+# Pipeline of dataset construction
+
+ROOT_FOLDER=$1
+
+if [ $# -eq 1]
+then
+    mkdir $ROOT_FOLDER;
+    echo "[[Start listing repositories.]]";
+    ./repo_lister.sh;
+    echo "[[Repository lists built.]]";
+    echo "[[Start cloning repositories to '$ROOT_FOLDER']]";
+    ./repo_clone.sh "$ROOT_FOLDER";
+    echo "[[Clone completed.]]";
+    echo "[[Start calculating ground truth.]]";
+    ./repo_arrange.sh "$ROOT_FOLDER";
+    echo "[[JSON ground truth calculated.]]";
+    echo "[[Start completing ground truth with text result.]]";
+    ./repo_arrange_text.sh "$ROOT_FOLDER";
+    echo "[[Ground truth completed.]]";
+    echo "[[Start arranging files by language.]]";
+    python3 ground_truth_arrange.py;
+    echo "[[Raw file dataset built in '$ROOT_FOLDER'.]]";
+else
+    echo "Please enter root folder correctly. Only one folder is needed.";
+fi