Scribd
Upload
12 views40 pages

Data Preprocessing - 241024 - 215531

Uploaded by

Alaa Shorbaji
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
12 views40 pages

Data Preprocessing - 241024 - 215531

Uploaded by

Alaa Shorbaji
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 40

Data Cleansing

&
Exploratory
Data Analysis

Created by
Seaborn Kusto
Official Team

Malka Trianto Maulana Fauzi Rizki

Rusyd Haryo Ishaq Wardah Afrinal

Abdussalam Nugroho Siregar Ali


Table of Data Cleansing

Content Exploratory Data Analysis


Data Cleansing
Presentations are communication tools that can be used as demonstrations, lectures, spee
Importing Libraries
We will start by importing the libraries we will require for data cleansing. These include Pandas, NumPy,
Matplotlib, and Seaborn
Pandas is a Python library for data analysis.
Numpy is a Python library for mathematical operations.
Matplotlib is a Python library for data visualization.
Seaborn is a Python library for data visualization and exploratory data analysis.
Uploading Dataset
The next step after importing libraries is uploading the dataset from the storage location to google colab.
The dataset that we will use is the Titanic.csv dataset obtained from Kaggle.

Reading Dataset
After the dataset is uploaded then we read the contents of the dataset.
Displaying DataFrame Information
The info() method prints information
about the DataFrame. The information
contains the number of columns,
column labels, column data types,
memory usage, range index, and the
number of cells in each column (non-
null values). Note: the info() method
actually prints the info.

We have to check the state of data in


each variable/column before taking any
action in data manipulation and data
cleaning
Column Age
The describe() method prints information about the data description starting with count, mean,
standard deviation, min, and max. The value_counts() method prints information counting for each
value. And the plot() method display graphs according to the format we specify
Column Age
because the Age coulumn shows skewness, so we
have to apply an imputation with median, and we
can check data information again.
Column Cabin
the sum of data entry is 891 yet the Cabin Column is 204. it means there must be NULL in the Cabin
column. and we can show Cabin column proportion

It showed that Cabin column value has too much unique


data and also the Cabin clumn info is not quite
informative to describe survived data. So we better
remove the Cabin column
Column Embarked
the sum of data entry is 891 yet the Embarked Column is 889. We can show the Embarked column
proportion. Embarked column is categorical data

we will apply an imputation on Embarked column. so we check the data type of the EMbarked
column first. Embarked column is categorical data so the imputation is using mode. from the
proportion of EMbarked column, S appeared the most. so S is the mode
After all data cleansing. we can check information data. There's no more missing value
Data Manipulation
Column SibSp and Column Parch
We will do data manipulation. Manipulation here doesn't mean changing the data value but to
ease a machine to read data. SibSp column (sibling Spouse) ia a column that state the number of
siblings or partner came with the pessenger. Parch(Parent Childern) column is a column that state
the number of parents or children came with the pessenger.
we will make a new column that shows whether the
pessenger is alone or coming with their family.
So we can show the new data
Data Visualization
Realtion between sex column and survived column
let's see the survived Sex column proportion and compare it with the Sex column which is not
survived. And we can show the visualization of Sex column which is surivived and not survived
Exploratory
Data Analysis
Exploratory Data Analysis, or EDA, is an important step in
any Data Analysis or Data Science project. EDA is the
process of investigating the dataset to discover patterns,
and anomalies (outliers), and form hypotheses based on
our understanding of the dataset.

EDA involves generating summary statistics for


numerical data in the dataset and creating various
graphical representations to understand the data better.
Before we delve into EDA, it is important to first get a sense of where EDA fits in the whole data
science process.
Importing Libraries
We will start by importing the libraries we will require for performing EDA. These include Pandas, NumPy,
Matplotlib, and Seaborn
Pandas is a Python library for data analysis.
Numpy is a Python library for mathematical operations.
Matplotlib is a Python library for data visualization.
Seaborn is a Python library for data visualization and exploratory data analysis.
Uploading Dataset
The next step after importing libraries is uploading the dataset from the storage location to google colab.
The dataset that we will use is the Titanic.csv dataset obtained from Kaggle.

Reading Dataset
After the dataset is uploaded then we read the contents of the dataset.
Displaying Top 5 Rows
Returns the top 5 rows of the dataset to have a look at how our dataset looks like.
Changing Index
Pandas default index starts from 0, while the dataset index of the PassengerId column starts from 1. Then
we will use the index dataset of column PassengerId.
Displaying DataFrame Information
The info() method prints information about the DataFrame. The information contains the number of
columns, column labels, column data types, memory usage, range index, and the number of cells in each
column (non-null values). Note: the info() method actually prints the info.
Checking Missing Value (NaN)
Checking whether there is a missing value (NaN) and also counting the number of the missing value in
each columns in the dataset.
Displaying Descriptive Statistics
Looking at descriptive statistic parameters for the dataset: Count, Mean, Standard Deviation, Maximum
and Minimum , Quartile (25%, 50% and 75%).
Displaying Unique Values in a Column
Displaying all of the unique value and its data types in a Column.
Displaying Proportion of Unique Values
Displays the data proportion of its unique values ​for the categoric data type.
Displaying Shape (Number of Rows and Columns)
Displays the data proportion of its unique values ​for the categoric data type.

Checking Duplicate Data


Checking the number of duplicate Data for each column.
Removing Duplicate Data
Remove all of duplicate data from the dataset.
Embarked Column
Embarked column has 2 null data on PassangerId numer 62 and 830
Embarked Column
Embarked is categoric data, we can use mode for imputation missing data in Embarked Column

Proportion Embarked
Before Imputation After Imputation
Embarked Column
Change the object data in Embraked ('S', 'C', 'Q') to numerical data (0, 1, 2)
Age Column
Data Titanic has 891 row, in Age Column only 714 row, its mean Age
Column has 177 missing data
Age Column
Age Colum has Skewness Distribution, because of that we can use
median for imputaion missing data
Age Column
Visualitation data Age Column

in this plot we can see the outliers, in this case the


outliers are passengers aged in range (0, +-5) and more
than +- 55
Sex Column
Data in Sex column only has two unique data, that is male and female . we want
to convert data object to data numerical

0 for male and 1 for female


Drop Data
Drop data is used when the existing data is uninformative and has a lot of unique data.,
in this case we use drop data in Cabin Column, Name Column and Ticket Column
Data Survived Visualitation
Proportion of Survived data
Data Survived Visualization
Making data frame from Survived data
Data Survived Visualization
Show chart using matplotlib.pyplot module
Data Survived Visualization
Show chart using seaborn module
Thank
You
Seaborn Kusto

reallygreatsite.com

You might also like