EDA-on-Google-PlayStore

EDA-on-Google-PlayStore

#impoting the library

import pandas as pd import numpy as np import seaborn as sns import matplotlib.pyplot as plt import warnings warnings.filterwarnings('ignore')

df = pd.read_csv("googleplaystore.csv") #data from csv file df.columns df.head() df.describe() #stats inform for numerical data #five point summary

df.duplicated().sum() df = df.drop_duplicates() df

df.dtypes

df['Content Rating'].dtype df['Reviews'] df['Reviews'].dtype type(df.Reviews) df.Reviews df[~df.Reviews.str.isnumeric()] df_copy = df.copy() df_copy

#restting the index as after dropping duplicates original index is still there

df_copy.reset_index(drop = True, inplace=True) df_copy df_copy[~df_copy.Reviews.str.isnumeric()] df_copy = df_copy.drop(df_copy.index[9990]) df_copy[~df_copy.Reviews.str.isnumeric()] df_copy['Reviews'] = df_copy['Reviews'].astype(int) df_copy.Reviews.dtype df_copy.info() df_copy["Size"].unique() item = '19M' item[-1] item.replace('M', '')

#1 MB = 1024 KB\

def size_process(item): if str(item)[-1] == 'M': res = float(str(item).replace('M', '')) res = res*1024 return res elif str(item)[-1] == 'k': res = float(str(item).replace('k', '')) return res else: return str(np.NaN)

df_copy['Size'] = df_copy['Size'].apply(size_process) df_copy.Size.dtype df_copy.Size = df_copy.Size.astype('float') df_copy.Size.dtype df_copy.info() df_copy['Installs'] df_copy['Installs'] = df_copy['Installs'].str.replace("+", "").str.replace(",", "") df_copy['Installs'].astype(int) df_copy["Price"].unique()

char_to_remove = ["+", ",", "$"] cols_to_remove = ["Installs", "Price"]

for char in char_to_remove: for col in cols_to_remove: df_copy[col] = df_copy[col].str.replace(char, "")

df_copy['Installs'] = df_copy['Installs'].astype(int) df_copy['Price'] = df_copy['Price'].astype(float)

df_copy['Last Updated'] df_copy['Last Updated'] = pd.to_datetime(df_copy['Last Updated']) df_copy['day'] = df_copy['Last Updated'].dt.day df_copy['month'] = df_copy['Last Updated'].dt.month df_copy['year'] = df_copy['Last Updated'].dt.year df_copy.dtypes df_copy['Android Ver'].unique() df_copy['Android Ver'] = df_copy['Android Ver'].str.replace("and up", "").str.replace("Varies with device","") df_copy['Android Ver'].unique() df_copy.App df[df.duplicated("App")]

df_copy = df_copy.drop_duplicates(subset = ["App"], keep = 'first') df_copy[df_copy.duplicated("App")] df_copy.dtypes df_copy.columns #EDA categorical_features = [feature for feature in df_copy.columns if df_copy[feature].dtype == 'O'] numerical_features = [feature for feature in df_copy.columns if df_copy[feature].dtype != 'O'] numerical_features df_copy[categorical_features] #categroical feature analysis df_copy["Type"].value_counts(normalize = True)*100

for col in categorical_features: print(f"{col} : {df_copy[col].value_counts(normalize = True)*100}")

df_copy["Android Ver"].value_counts(normalize = True)*100 df_copy["Type"].value_counts(normalize = True)*100

#Q. Which category is the most popular category in the app?

#pie chart

df_copy["Category"].value_counts().plot.pie(y = df["Category"], figsize = (12, 12), autopct = '%1.1f%%')

#family is the most popular category with 19 % of share #Q. what is the top 10 important category cat = df_copy["Category"].value_counts()[:10] category = cat.reset_index() category.columns = ["Groups", "count"] sns.barplot(category, x = category['Groups'], y = category['count'])

#data from google playstore csv file