Python-сообщество

 >>> processed_text = sample_text.apply(lambda x: re.findall('[a-zA-Z]{5,}', x))  

 6    [Plasticity, Mediated, Competitive, Learning]
7    [Morphology, Classification, using, Analogue]

 6    Plasticity Mediated Competitive Learning
7    Morphology Classification using Analogue

 >>> text = 'Hello this is the^& test &%text with some*&* digits 12449/565)^%'
>>> preprocessed_text = re.sub('[^a-zA-Z]+', ' ', text)
>>> print(preprocessed_text)
'Hello this is the test text with some digits'

 >>> result = re.sub(r'\W*\b\w{1,4}\b', '', preprocessed_text.lower())
>>> print(result)
'hello digits'