Ex-03EDA

AIM

To perform EDA on the given data set.

Explanation

The primary aim with exploratory analysis is to examine the data for distribution, outliers and anomalies to direct specific testing of your hypothesis.

ALGORITHM

STEP 1

Import the required packages(pandas,numpy,seaborn).

STEP 2

Read and Load the Dataset.

STEP 3

Remove the null values from the data and remove the outliers.

STEP 4

Remove the non-numerical data columns using drop() method.

STEP 5

Return object containing counts of unique values using (value_counts()).

STEP 6

Plot the counts in the form of Histogram or Bar Graph.

STEP 7

Find the pairwise correlation of all columns in the dataframe(.corr()).

STEP 8

Save the final data set into the file.

CODE

Program 
Developed by: Dhivya Shri.B
Register no:212221230009

import pandas as pd
import numpy as np
import seaborn as sns
df=pd.read_csv("titanic_dataset.csv")
df

#removing data containing too many null values
df.isnull().sum()
df.drop('Cabin',axis=1,inplace=True)

#analysing dataframe contents
df.info()
df.isnull().sum()

#cleaning data
df['Age']=df['Age'].fillna(df['Age'].median())
df['Embarked']=df['Embarked'].fillna(df['Embarked'].mode()[0])
df.isnull().sum()

#data is cleaned, checking for outliers
df.boxplot()
#removing outliers
cols = ['Age', 'Fare','SibSp','Parch','Fare']
Q1 = df[cols].quantile(0.25)
Q3 = df[cols].quantile(0.75)
IQR = Q3 - Q1
df = df[~((df[cols] < (Q1 - 1.5 * IQR)) |(df[cols] > (Q3 + 1.5 * IQR))).any(axis=1)]
df
df.boxplot()

#maximum outliers removed
### Performing data analysis
#statistical analysis for single data group
df['Survived'].value_counts()
df['Pclass'].value_counts()
df['SibSp'].value_counts()
df['Embarked'].value_counts()
df['Sex'].value_counts()

#statistical analysis for two data groups
pd.crosstab(df["Pclass"],df["Survived"])
pd.crosstab(df["Sex"],df["Survived"])
pd.crosstab(df["Embarked"],df["Survived"])
pd.crosstab(df["SibSp"],df["Survived"])

#graphical analysis of categorical data--univariate
sns.countplot(x="Survived",data=df)
sns.countplot(x="Pclass",data=df)
sns.countplot(x="SibSp",data=df)
sns.countplot(x="Embarked",data=df)
sns.countplot(x="Sex",data=df)

#graphical analysis of non-categorical data or data with multiple categories--univariate
sns.displot(df["Fare"])
sns.displot(df["Age"])

#graphical analysis of categorical data--bivariate
sns.countplot(x="Sex",hue="Survived",data=df)
sns.countplot(x='Pclass',hue='Survived',data=df)
sns.countplot(x='Embarked',hue='Survived',data=df)

#graphical analysis of non-categorical data or data with multiple categories--bivariate
sns.displot(df[df["Survived"]==0]["Age"])
sns.displot(df[df["Survived"]==1]["Age"])

#Graphical representation of data--multivariate 
df.drop('Parch',axis=1,inplace=True)
sns.heatmap(df.corr(),annot=True)

#analysing pairwise correlation of columns in dataset
df.corr()