Privacy related algorithm in Spark

Question

I have the following code on k-Anonymity ('*' suppress values version) that I want to check its correctness. The code compiles and outputs as expected.

public class KAnonymity {
    /**
     * @param k_anonymity_constant k constant
     * @param qidAtts              list of Quasi-ID attributes
     * @param initDataset          the dataset to k-Anonymize
     * @return k-Anonymized Datasets
     */
    public static Dataset<Row> kAnonymizeBySuppression( Dataset<Row> initDataset, List<String> qidAtts, Integer k_anonymity_constant) {
        List<Column> groupByQidAttributes = qidAtts.stream().map(functions::col).collect(Collectors.toList());
        Dataset<Row> processedDataset = initDataset.withColumn("qidsFreqs", count("*").over(Window.partitionBy(groupByQidAttributes.toArray(new Column[groupByQidAttributes.size()]))));
        for (String qidAtt : qidAtts)
            processedDataset = processedDataset.withColumn(qidAtt, when(col("qidsFreqs").leq(lit(k_anonymity_constant)), lit('*')).otherwise(col(qidAtt)));
        return processedDataset.drop("qidsFreqs");

    }
}

PS: Unit test case (k=5)

//Read the csv dataset 
Dataset<Row> dataset = spark.read().csv(pathToCSV);
// Anonymize it
List<String qidAttsList = new ArrayList<>();
qidAttsList.add("qidAtt1");
qidAttsList.add("qidAtt2");
...
 Dataset<Row> anonymizedDF = KAnonymity.kAnonymizeBySuppression(dataset, qidAttsList, 5);