Permalink
Cannot retrieve contributors at this time
Name already in use
A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?
binary-classification-of-tweets-about-protest/Untitled.ipynb
Go to fileThis commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
236 lines (236 sloc)
9.07 KB
This file contains 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
{ | |
"cells": [ | |
{ | |
"cell_type": "code", | |
"execution_count": null, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"from pandas import read_csv, DataFrame\n", | |
"from sklearn.model_selection import train_test_split\n", | |
"import language_tool_python\n", | |
"\n", | |
"#import textstat\n", | |
"#from imblearn.over_sampling import RandomOverSampler\n", | |
"from sklearn.preprocessing import MinMaxScaler\n", | |
"from sklearn.metrics import accuracy_score\n", | |
"from sklearn.linear_model import LogisticRegression\n", | |
"from xgboost import XGBClassifier\n", | |
"from sklearn.svm import SVC\n", | |
"import pandas as pd\n", | |
"from spellchecker import SpellChecker\n", | |
"\n", | |
"from yellowbrick.cluster import KElbowVisualizer\n", | |
"\n", | |
" \n", | |
"\"\"\"\n", | |
"# Mention the language keyword \n", | |
"tool = language_check.LanguageTool('en-US') \n", | |
"def get_data(df,sampling_method=\"None\"):\n", | |
" dataset = df.values\n", | |
"\n", | |
" X = dataset[:, 1:]\n", | |
" Y = dataset[:,0]\n", | |
" seed = 7\n", | |
" test_size = 0.33\n", | |
" train_size = 0.67\n", | |
"\n", | |
" oversample = RandomOverSampler(sampling_strategy='all')\n", | |
" X, Y = oversample.fit_resample(X,Y)\n", | |
"\n", | |
" X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=test_size, random_state=seed, train_size=train_size)\n", | |
" calc(X_train,X_test,Y_train, Y_test)\n", | |
"def calc( X_train,X_test,Y_train,Y_test):\n", | |
" best_model = XGBClassifier(\n", | |
" min_child_weight=1,\n", | |
" gamma=0.1,\n", | |
" subsample=1.0,\n", | |
" colsample_bytree=1.0,\n", | |
" max_depth=20,\n", | |
" eta=1)\n", | |
" \n", | |
" best_model.fit(X_train,Y_train)\n", | |
"\n", | |
" y_pred = best_model.predict(X_test)\n", | |
" predictions = [round(value) for value in y_pred]\n", | |
" #accuracy = round(accuracy_score(Y_test, predictions) * 100.0,3)\n", | |
" accuracy = accuracy_score(Y_test, y_pred)\n", | |
" print(accuracy)\n", | |
"\"\"\"\n", | |
"tool = language_tool_python.LanguageTool('en-US')\n", | |
"file_path = 'data/extracted/normalized/2-gram-4-clusters.csv'\n", | |
"spell = SpellChecker()\n", | |
"preprocessed = read_csv('data/preprocessed/april-21.csv')\n", | |
"df = read_csv('data/raw/april-21.csv',encoding='ISO-8859-1')\n", | |
"small_df = df\n", | |
"def process(text):\n", | |
" #spell.unknown(text.split(\" \"))\n", | |
" try:\n", | |
" \n", | |
" return len(tool.check(text))\n", | |
" except:\n", | |
" print(\"ERROR\")\n", | |
" print(text)\n", | |
" return 0\n", | |
"\n", | |
"\n", | |
"small_df['spelling_errors'] = small_df['text'].apply(process)\n", | |
"yes_df = small_df[small_df['Y/N'] == 'Y']\n", | |
"no_df = small_df[small_df['Y/N'] == 'N']\n", | |
"print(yes_df['spelling_errors'].mean())\n", | |
"print(no_df['spelling_errors'].mean())\n", | |
"\n", | |
"\n", | |
"\n", | |
"\n", | |
"\n", | |
"#df['readability'] = df['text'].apply(lambda x: textstat.flesch_reading_ease(x))\n", | |
"\n", | |
"\n", | |
"\n", | |
"\n" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 8, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"data": { | |
"text/plain": [ | |
"'df[\\'word_count\\'] = df[\\'text\\'].apply(lambda x : len(x.split()))\\ndf[\\'char_count\\'] = df[\\'text\\'].apply(lambda x : len(x.replace(\" \",\"\")))\\ndf[\\'total_length\\'] = df[\\'text\\'].apply(len)\\n\\n\\nfor each in [\\'word_count\\',\\'char_count\\',\\'total_length\\']:\\n print(yes_df[each].mean())\\n print(no_df[each].mean())\\n \\ndf[\\'hashtag_count\\'] = df[\\'hashtags\\'].apply(lambda x: len(x.split(\" \")) if isinstance(x,str) else 0)\\ndf[\\'num_unique_words\\'] = df[\\'text\\'].apply(lambda x: len(set(w for w in x.split())))\\ndf[\\'capitals\\'] = df[\\'text\\'].apply(lambda comment: sum(1 for c in comment if c.isupper()))\\ndf[\\'num_exclamation_marks\\'] =df[\\'text\\'].apply(lambda x: x.count(\\'!\\'))\\ndf[\\'num_question_marks\\'] = df[\\'text\\'].apply(lambda x: x.count(\\'?\\'))\\ndf[\\'num_punctuation\\'] = df[\\'text\\'].apply(lambda x: sum(x.count(w) for w in \\'.,;:\\'))\\ndf[\\'num_symbols\\'] = df[\\'text\\'].apply(lambda x: sum(x.count(w) for w in \\'*&$%\\'))\\n\\n\\n\\nfor each in [\\'hashtag_count\\',\\'num_unique_words\\',\\'capitals\\',\\'num_exclamation_marks\\',\\'num_question_marks\\',\\'num_punctuation\\',\\'num_symbols\\']:\\n print(each)\\n print(\"SUPPORT:\", yes_df[each].mean())\\n print(\"NOT SUPPORITING: \",no_df[each].mean())\\n '" | |
] | |
}, | |
"execution_count": 8, | |
"metadata": {}, | |
"output_type": "execute_result" | |
} | |
], | |
"source": [ | |
"\n", | |
"\n", | |
"\n", | |
"\n", | |
"\"\"\"df['word_count'] = df['text'].apply(lambda x : len(x.split()))\n", | |
"df['char_count'] = df['text'].apply(lambda x : len(x.replace(\" \",\"\")))\n", | |
"df['total_length'] = df['text'].apply(len)\n", | |
"\n", | |
"\n", | |
"for each in ['word_count','char_count','total_length']:\n", | |
" print(yes_df[each].mean())\n", | |
" print(no_df[each].mean())\n", | |
" \n", | |
"df['hashtag_count'] = df['hashtags'].apply(lambda x: len(x.split(\" \")) if isinstance(x,str) else 0)\n", | |
"df['num_unique_words'] = df['text'].apply(lambda x: len(set(w for w in x.split())))\n", | |
"df['capitals'] = df['text'].apply(lambda comment: sum(1 for c in comment if c.isupper()))\n", | |
"df['num_exclamation_marks'] =df['text'].apply(lambda x: x.count('!'))\n", | |
"df['num_question_marks'] = df['text'].apply(lambda x: x.count('?'))\n", | |
"df['num_punctuation'] = df['text'].apply(lambda x: sum(x.count(w) for w in '.,;:'))\n", | |
"df['num_symbols'] = df['text'].apply(lambda x: sum(x.count(w) for w in '*&$%'))\n", | |
"\n", | |
"\n", | |
"\n", | |
"for each in ['hashtag_count','num_unique_words','capitals','num_exclamation_marks','num_question_marks','num_punctuation','num_symbols']:\n", | |
" print(each)\n", | |
" print(\"SUPPORT:\", yes_df[each].mean())\n", | |
" print(\"NOT SUPPORITING: \",no_df[each].mean())\n", | |
" \"\"\"\n", | |
"\n", | |
"\n" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 5, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"df = pd.read_csv('data/preprocessed/without_stop_words.csv', encoding='ISO-8859-1')\n", | |
"\n", | |
"def count_if_exist(corpus, word):\n", | |
" word_count = corpus.get(word,0) + 1\n", | |
" corpus[word] = word_count\n", | |
"\n", | |
"def convert_dict_to_csv(input_dict,name):\n", | |
" pd.DataFrame(input_dict.items()).to_csv(\"./data/computed/{name}.csv\".format(name=name),header=None,index=None)\n", | |
"general_word_count = {}\n", | |
"supporting_word_count = {} \n", | |
"non_supporting_word_count = {}\n", | |
"word_support_probability = {}\n", | |
"for i in range(1,len(df)):\n", | |
" cur_row = df.iloc[i]\n", | |
" words = cur_row.text.split(' ')\n", | |
" for word in words:\n", | |
" count_if_exist(general_word_count,word)\n", | |
" if(cur_row['Y/N'] == 'Y'):\n", | |
" count_if_exist(supporting_word_count,word)\n", | |
" else:\n", | |
" count_if_exist(non_supporting_word_count,word)\n", | |
"\n" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 28, | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"name": "stdout", | |
"output_type": "stream", | |
"text": [ | |
"['today', 'say', 'stay', 'take', 'like', 'amp', 'protester', 'gridlock', 'think', 'many', 'michigan', 'make', 'live', 'operationgridlock', 'need', 'people', 'get', 'protest', 'right', 'governor', 'state', 'would', 'order', 'one', 'home', 'trump', 'want', 'see', 'work', 'go', 'u', 'mi', 'lansing']\n" | |
] | |
} | |
], | |
"source": [ | |
"supporting_pair = sorted(supporting_word_count.items(),key=lambda x: x[1],reverse=True)\n", | |
"non_supporting_pair = sorted(non_supporting_word_count.items(),key=lambda x: x[1],reverse=True)\n", | |
"supporting_words = list(map(lambda x: x[0], supporting_pair))[:50]\n", | |
"non_supporting_words = list(map(lambda x: x[0], non_supporting_pair))[:50]\n", | |
"intersection = list(set(supporting_words) & set(non_supporting_words))\n", | |
"print(intersection)" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": 22, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"def convert_dict_to_csv(input_dict,name):\n", | |
" pd.DataFrame(input_dict.items()).to_csv(\"./data/computed/{name}.csv\".format(name=name),header=['count'],index=['readability'])" | |
] | |
}, | |
{ | |
"cell_type": "code", | |
"execution_count": null, | |
"metadata": {}, | |
"outputs": [], | |
"source": [ | |
"pd" | |
] | |
} | |
], | |
"metadata": { | |
"kernelspec": { | |
"display_name": "Python 3", | |
"language": "python", | |
"name": "python3" | |
}, | |
"language_info": { | |
"codemirror_mode": { | |
"name": "ipython", | |
"version": 3 | |
}, | |
"file_extension": ".py", | |
"mimetype": "text/x-python", | |
"name": "python", | |
"nbconvert_exporter": "python", | |
"pygments_lexer": "ipython3", | |
"version": "3.6.9" | |
} | |
}, | |
"nbformat": 4, | |
"nbformat_minor": 4 | |
} |