Created
November 19, 2016 16:54
-
-
Save 0x7067/7c95df3ef3c313c4af3011a81dd38163 to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| [u'psic\xf3loga', u'hoje', u'deu', u'vontade', u'ler', u'romancezinho', u'bem', u'self', u'insert', u'esquecer', u'existo'] | |
| [] | |
| [u'whoa', u'must', u'biggest', u'windows', u'blue', u'screen', u'death', u'ever', u'seen'] | |
| [u'whenever', u'see', u'someone', u'write', u'would'] | |
| [u'equipe', u'rocket', u'destruindo', u'padr\xf5es', u'g\xeanero', u'desde'] | |
| [u'want', u'relationship', u'even', u'though', u'women', u'shepard', u'miga', u'mina', u'\xe9', u'_azul_', u'talvez', u'atente', u'pra', u'antes', u'g\xeanero'] | |
| [u'adoro', u'\xe9', u'fucking', u'wrpg', u'ainda', u'assim', u'momento', u'b-but', u'girls', u'liara'] | |
| [] | |
| [u'mass', u'effect'] | |
| [u'cena', u'ser', u'maravilhosa', u'nunca', u'mds', u'<3'] | |
| PS: Run filter_tweets.py with "python filter_tweets.txt >> after.txt" |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| psicóloga hoje me deu vontade de ler um romancezinho bem self insert e esquecer que eu existo. | |
| rt AT_USER URL | |
| rt AT_USER whoa! "this must be the biggest windows blue screen of death ever seen": URL | |
| rt AT_USER me whenever i see someone write "should of" or "would of" URL | |
| rt AT_USER equipe rocket destruindo padrões de gênero desde 1997 URL | |
| you want a relationship with me? even though we're both women?" shepard, miga, a mina é _azul_. talvez atente pra isso antes do gênero | |
| adoro que é um fucking wrpg e ainda assim tem o momento "b-but we're both girls!" com a liara | |
| AT_USER s | |
| AT_USER mass effect | |
| essa cena não para de ser maravilhosa nunca mds <3 |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| #!/usr/bin/env python | |
| # -*- coding: utf-8 -*- | |
| import json | |
| import re | |
| import string | |
| from collections import Counter | |
| from nltk.tokenize import TweetTokenizer | |
| from nltk.corpus import stopwords | |
| def normalize_contractions(tokens): | |
| """Example of normalization for English contractions. | |
| Return: generator | |
| """ | |
| token_map = { | |
| "i'm": "i am", | |
| "you're": "you are", | |
| "it's": "it is", | |
| "we're": "we are", | |
| "we'll": "we will", | |
| "vc" : "voce", | |
| "pq" : "porque" | |
| } | |
| for tok in tokens: | |
| if tok in token_map.keys(): | |
| for item in token_map[tok].split(): | |
| yield item | |
| else: | |
| yield tok | |
| def process(text, tokenizer=TweetTokenizer(), stopwords=[]): | |
| """Process the text of a tweet: | |
| - Lowercase | |
| - Tokenize | |
| - Stopword removal | |
| - Digits removal | |
| Return: list of strings | |
| """ | |
| tokenizer = TweetTokenizer(strip_handles=True, reduce_len=True) | |
| text = text.lower() | |
| tokens = tokenizer.tokenize(text) | |
| # If we want to normalize contraction, uncomment this | |
| tokens = normalize_contractions(tokens) | |
| return [tok for tok in tokens if tok not in stopwords and not tok.isdigit()] | |
| tknzr = TweetTokenizer(preserve_case=False, strip_handles=True, reduce_len=True) | |
| punct = list(string.punctuation) | |
| custom_list = ['belo', 'horizonte', 'mg', 'at_user', 'url'] | |
| stopword_list = stopwords.words('portuguese') + stopwords.words('english') + punct + ['rt', 'via'] + custom_list | |
| i = open('tweets_pre_processed.txt', 'r') | |
| line = i.readline() | |
| while line: | |
| print(process(text=line, tokenizer=tknzr, stopwords=stopword_list)) | |
| line = i.readline() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment