from nltk.corpus import stopwords stop_words = set(stopwords.words('english')) df['n'] = df['text'].str.split().apply(lambda x: len(set(x) & stop_words))