“…From this, a dictionary comprising all the unique words can be generated. We have used the count vectorizer and term frequency-inverse document frequency (TF-IDF) for feature extraction to determine first the most mentioned words [44] for each language and later the main topics.…”