A MINI PROJECT REPORT ON

“Big Mart Sales Prediction”

Submitted by
Kaushal Chande (17IT1012)
Makarand Madhavi (17IT1020)
Omkar Chalke (17IT2030)
Dnyanisha Gondhalekar (17IT1037)
Siddhi Kale (17IT2020)
Tanvi Chavan (17IT2025)

Under The Guidance Of

Prof. Reshma Gulwani

Department of Information

Technology

Ramrao Adik Institute of

Technology, Nerul, Navi Mumbai

(Affiliated to University of Mumbai)\

2020
 CERTIFICATE
This is to certify that the project entitled 'Big Mart Sales
Prediction' being submitted by Kaushal Chande (17IT1012),
Makarand Madhavi (17IT1020), Omkar Chalke (17IT2030),
Dnyanisha Gondhalekar (17IT1037), Siddhi Kale (17IT2020),
Tanvi Chavan (17IT2025) to the University of Mumbai in partial
fulfilment of the requirement for the award of the degree Of ‘T.E.
I.T’ in “BUSINESS INTELLIGENCE LAB”.

Project Guide External Examiner Head of Department

(Prof. Reshma Gulwani) ( ) (Dr. Ashish Jadhav)
 DECLARATION

We declare that this written submission represents our ideas

in our own words and where others ideas or words have been
included, we have adequately cited and Referenced the
original sources. We also declare that we have adhered to all
principles Of academic honesty and integrity and have not
misrepresented or fabricated or Falsified any
idea/data/fact/source in our submission. We understand that
any violation Of the above will be cause for disciplinary
action by the Institute and can also evoke Penal action from
the sources which have thus not been properly cited or from
whom Proper permission has not been taken when needed.

1. Kaushal Chande (17IT1012) ( )

2. Makarand Madhavi (17IT1020) ( )

3. Omkar Chalke (17IT2030) ( )

4. Dnyanisha Gondhalekar (17IT1037) ( )

5. Siddhi Kale (17IT2020) ( )

6. Tanvi Chavan (17IT2025) ( )

Date:

Place:
 ACKNOWLEDGEMENT

The project “Big Mart Sales Prediction” is creative work of many

minds. A proper synchronization between individual is must for any
project to be completed successfully. One can not imagine the power
of the force that guides us all and neither can we succeed without
acknowledging it.
We would like to express our gratitude to Principal Dr. Mukesh D.
Patil and Dr. Ashish Jadhav, our Head of the department,
Information Technology Engineering for encouraging and Inspiring us
to carry out the project in the department lab.
We would also like to thank our Guide Prof. Reshma Gulwani
Department of the Information technology engineering for her expert
guidance, encouragement and valuable suggestions at every step.
We also would like to thank all the staff members Department of the
Information Technology Engineering for providing us with the
required facilities and support towards the Completion of the project.
Last but not the least we are thankful to our parents and friends for
their constant Inspiration, encouragement and well wishes by which we
have made a challenging project.

Kaushal Chande (17IT1012)

Makarand Madhavi (17IT1020)

Omkar Chalke (17IT2030)

Dnyanisha Gondhalekar (17IT1037)

Siddhi Kale (17IT2020)

Tanvi Chavan (17IT2025)

 PREFACE
We take great opportunity to present this Mini Project report
on “Big Mart Sales Prediction” and put before readers
some useful information regarding our project. We have
made sincere attempts and taken every care to present this
matter in precise and compact form, the language being as
simple as possible. We are sure that the information
contained in this volume certainly prove useful for better in
sight in the scope and dimension of this project in it true
perspective. The task of the completion of the project
though being difficult was made quite simple, Interesting
and successful due to deep involvement and complete
dedication of our group members.
 TABLE OF CONTENTS

Declaration .................................................................. I

Acknowledgement… ....................................................II

Preface…......................................................................III

Table of Contents ....................................................... IV

 TABLE OF CONTENTS

1. PROBLEM STATEMENT .......................................................8

2. PROPOSED SYSTEM ............................................................. 9

3. DATASET .............................................................................10

4. PLATFORM USED ............................................................... 11

5. IMPLEMENTATION DETAIL ................................................ 12

6. RESULT ..............................................................................18

7. BUSINESS INTELLIGENCE .................................................. 21

8. CONCLUSION .................................................................... 22
 1. PROBLEM STATEMENT

The data scientists at Big Mart have collected the sales data for 1559 products
across 10 stores in different cities. Also, certain attributes of each product and
store have been defined. The aim is to build a predictive model and find out
the sales of each product at a particular store which would help the sales team
to plan financial growth and adopt suitable production policy.
 2. PROPOSED SYSTEM

The goal of the Big Mart sales prediction is to build a regression model to
predict the sales of each of new products in each of the 10 different Big Mart
outlets. This model helps Big Mart understand the properties of products and
stores that play an important role in increasing their overall sales.
 3. DATASET

We have used Big Mart Sales Dataset containing twelve attributes. Six of
them give data related to product and remaining six attributes give data
related to outlets.
The attributes and their description is as follows :

• Item Identifier - Unique Product ID

• Item Weight - Weight of the Product
• Item Fat Content - Whether the product is low fat or not
• Item Visibility - The percentage of total display area of all products in a
store to the particular product
• Item Type - The category to which the product belongs
• Item MRP - Maximum Retail Price (list price) of the product
• Outlet Identifier - Unique Store ID
• Outlet Establishment Year - The year in which store was established
• Outlet Size - The size of the store in terms of ground area covered
• Outlet Location Type - The type of city in which the store is located
• Outlet Type - Whether the outlet is just a grocery store or some sort of
supermarket
• Item_Outlet_Sales - Sales of the product in a particular store. This is the
outcome variable to be predicted.
Fig 3.0 - Dataset

Fig 3.1 - Dataset

 4. PLATFORM

1. We have used Python Jupyter Notebook as a platform for

implementation, coding and output.

2. We took Big Mart Dataset from Kaggle.

3. We have used four Python Libraries :

i. Pandas – Importing and Exporting data and Data Manipulation

ii. Seaborn – Data Visualization
iii. Sklearn – Model Making
iv. Matplotlib – Graphs
 5. IMPLEMENTATION

Data Exploration:
In this we have explored the dataset deeply by plotting graphs of
various attributes. By executing this we got an idea about which are
the major attributes will help us in finding the result.

Data Cleaning:
In this the data is cleaned by removing NULL values.

Data Transformation:
In this the data is transformed into Consistent data so that we can use
it easily while finding the results.

Train Model:
We have used the Train Dataset to train the model. The technique
used for training is "Linear Regression".

Test Model:
the Test Dataset is used for Testing Purpose and for finding the result.
After Testing the data we got accuracy about 79%.

Fig 5.0 – Implementation flow

5.1 Data Exploration
With the help of various graphs and Data visualization tools the dataset is
explored to understand attributes and relations.

Fig 5.1 – Graph showing importance on outlet type on sales

Fig 5.2 – Heatmap showing correlation between every pair of numeric attributes
 Fig 5.3 Fig 5.4
(count of outlet size) (impact of location type on sales)

Fig 5.5 distribution of fat content over sales

Fig 5.6 impact of visibility on item sales

5.2 Data Cleaning
Data is cleaned to replace null values and make values
consistent.
 5.3 Data Transformation
Data is transformed to make it suitable to be trained for a regression
model.

Mathematical Operation

Combining Data

Dummy Columns
 5.4 Model Making
Training dataset is split into two parts; Independent variables (X_train)
consisting of item and outlet variables, and dependent variable (Y_train)
i.e sales

Linear Regression model is prepared with the help of sklearn library

 6. RESULT

Independent variables from testing dataset are passed into the model to predict
the sales and result is obtained

The model gave an accuracy of 79.36%

 6.1 Result Analysis

The predicted sales obtained can be further analyzed to make suitable business decisions

Fig 5.6 shows predicted sales of each outlet

Fig 5.6 shows sales distribution of each item type

 Result Analysis

Fig 5.6 shows predicted mean sales by outlet type

 7. HOW BUSINESS INTELLIGENCE IS TO BE USED

Big Mart Sales Prediction provides visibility into which products are selling
the most in the market. Production team can identify areas of improvement
for low-selling products. Sales team can identify the geographical areas of
importance for selling the products. Sales team can adopt suitable production
policy so that the problem of overproduction or shortage can be avoided. If
there is an increase in sales of a product means that the demand for that
product has increased, this helps sales team to plan their supply to meet the
increased demand.
 CONCLUSION

Here, after successfully completing this project well before

time we learned the art of working in a team with proper
coordination among team members. The difficulties
occurred during making of this project was tackled by the
teammates. The enthusiasm of every member involved in
this project made it possible to achieve this goal. We hope
that this project is really going to be helpful to all the people.