import pandas as pd import numpy as np import matplotlib.pyplot as plt %matplotlib inline import seaborn as sns print('Setup completed!!')

data = pd.read_csv('/work/Womens Clothing E-Commerce Reviews.csv') data.head(10)

data.dtypes

data.drop(['Unnamed: 0','Clothing ID'], axis=1, inplace=True)

data.rename(columns={'Review Text': 'text', 'Positive Feedback Count': 'feedback_count', 'Division Name': 'Division', 'Department Name': 'Department', 'Class Name':'Class', 'Recommended IND': 'Recommended'},inplace=True)

print(data.columns)

data.isnull().sum()

data.dropna(axis=0, inplace=True)

data.isnull().sum()

sns.histplot(data=data, x='Age', kde=True) plt.show()

sns.histplot(data=data, x='Rating', kde=False)

sns.set_style('white') sns.barplot(x=data['Rating'], y=data['Class'],palette="Blues_d")

sns.set_theme(style="white") sns.color_palette("flare", as_cmap=False) sns.barplot(x=data['Age'], y=data['Class'],palette="Reds_d")

sns.displot(data=data, x='Rating', hue="Class", kind="kde", height=6, multiple="fill", clip=(0, None), )

sns.boxplot(x=data['Department'], y=data['Age']) plt.show()

data.groupby(['Department']).sum().plot(kind='pie', subplots=False, shadow = False,startangle=90,figsize=(15,10), y='Rating')

sns.countplot(data=data, y='Class', hue='Recommended', palette = "Set1")

sns.countplot(data=data, y='Department', hue='Recommended' ,palette = "Set1")

sns.countplot(data=data, y='Division', hue='Recommended', palette = "Set1")

import nltk from nltk.corpus import stopwords from nltk.stem.porter import PorterStemmer

data.columns

data['all_text'] = data['text']

nltk.download('punkt');

def tokenize(column): # Tokenize a Pandas dataframe columns and returns a list of tokens. tokens = nltk.word_tokenize(column) return [w for w in tokens if w.isalpha]

data['tokenized'] = data.apply(lambda x: tokenize(x['all_text']), axis=1) data[['text', 'tokenized']].head()

nltk.download('stopwords');

def remove_stopwords(tokenized_column): # this fuction will retun a list of tokens with English stopwords removed stops = set(stopwords.words('english')) return [word for word in tokenized_column if not word in stops]

data['stopwords_removed'] = data.apply(lambda x: remove_stopwords(x['tokenized']), axis=1) data[['text', 'stopwords_removed']].head()

def apply_stemming(tokenized_column): # this function will return a list of tokens with PorterStemming applied stemmer = PorterStemmer() return [stemmer.stem(word) for word in tokenized_column]

data['porter_stemmed'] = data.apply(lambda x: apply_stemming(x['stopwords_removed']), axis=1) data[['text', 'porter_stemmed']].head()

def rejoin_words(tokenized_column): # this fuction will rejoin the tokenized words list into a single string return (" ".join(tokenized_column))

data['rejoined'] = data.apply(lambda x: rejoin_words(x['porter_stemmed']), axis=1) data[['text', 'rejoined']].head()