diff --git a/Jenkinsfile b/Jenkinsfile
index cc731a5..fabcd48 100644
--- a/Jenkinsfile
+++ b/Jenkinsfile
@@ -31,8 +31,16 @@ node {
               "KAGGLE_KEY=${params.KAGGLE_KEY}" ]) {
          sh 'kaggle datasets download -d mssmartypants/water-quality > output.txt'
          sh 'unzip -o water-quality.zip >> output.txt'
-         sh 'head -n ${params.CUT} waterQuality1.csv > temp.csv'
-         sh 'mv temp.csv waterQuality1.csv'
+         def csvFile = new File('waterQuality1.csv')
+         def lines = csvFile.readLines()
+         def cut = ${params.CUT}.toInteger()
+         def randomLines = new Random().with { random ->
+            (0..lines.size() - 1).sort { random.nextInt() }[0..(cut - 1)].collect { lines[it] }
+         }
+         def tempFile = new File('temp.csv')
+         tempFile.write(randomLines.join('\n'))
+         csvFile.text = tempFile.text
+         tempFile.delete()
          sh 'python3 create_dataset.py >> output.txt'
          archiveArtifacts artifacts: 'waterQuality.csv, output.txt'
      }