I felt the same about number 3. [x in tweet.split() for x in wordlist] Seems better to me than map(lambda x: x in tweet.split(),wordlist) Although the n

I felt the same about number 3. [x in tweet.split() for x in wordlist] Seems better to me than map(lambda x: x in tweet.split(),wordlist) Although the natural thing for me to do would be [x for x in wordlist if x in tweet] Which both generates a list of words instead of booleans(['scala', 'sbt'] instead of [True, False, False, False, False]) and also matches the 'sbt' in tweet.