Hands-On Machine Learning for Algorithmic Trading

Named entity recognition

spaCy enables named entity recognition using the .ent_type_ attribute:

for t in sentences[0]:    if t.ent_type_:        print('{} | {} | {}'.format(t.text, t.ent_type_, spacy.explain(t.ent_type_)))annual | DATE | Absolute or relative dates or periodsthe | DATE | Absolute or relative dates or periodsyear | DATE | Absolute or relative dates or periods

textacy facilitates access to the named entities that appear in the first article:

from textacy.extract import named_entitiesentities = [e.text for e in named_entities(doc)]pd.Series(entities).value_counts()year                          4US                            2South-East Asia Earthquake    2annual                        2Tsunami Blog                  2