Lab 2.2 last Q SRC

from sussex_nltk.corpus_readers import MedlineCorpusReader, ReutersCorpusReader, TwitterCorpusReader

readers = [MedlineCorpusReader(), ReutersCorpusReader(), TwitterCorpusReader()]
sample_size = 10000

for reader in readers:
    sentences = map(lambda s: word_tokenize(s.lower()), reader.sample_raw_sents(sample_size))
    stops = 0
    others = 0

    for tokens in sentences:
        for token in tokens:
            if token in stop:
                stops += 1
            else:
                others += 1
    print(f"{round(stops / (stops + others) * 100, 2)}% were stop words with a total of {stops}/{stops + others}")