INNER CODE UNIT · Python

_generate_stop_words

vgrabovets/multi_rake · multi_rake/algorithm.py:102

    def _generate_stop_words(self, text):
        stop_words = set()

        text = keep_only_letters(text)

        if not text:
            return stop_words

        text = text.split()

        word_counts = Counter(text).most_common()
        counts_sample = [word_count[1] for word_count in word_counts]

        upper_bound = np.percentile(
            counts_sample,
            self.generated_stopwords_percentile,
        )

View source record →