ABESEC Ghaziabad

Department of Computer Science & Engineering

SYNOPSIS REPORT
(2019-20)

Project Title: Sales Prediction and Analysis using Regression

Project Type Application and Research

Name Roll Number Section Signature

Group member (1) Prashant Gupta 1603210132 C

Group member (2) Pradumn Kumar Singh 1603210127 C
Group member (3) Kundan Kumar 1603210098 C
Project Guide Ms. Nidhi Singh Remarks:

Signature

Date of
submission

1.1. Problem Introduction

1.1.1. Motivation
A firm’s success is determined by its sales and the performance of its product in the industry is
also identified by its sales. Hence in order to improve the standard of the firm /concern the
strategies, the techniques the methodologies are built. In this process of development forecasting
of certain key factors it may pave the way for the successful venture of the concern

1.1.2. Project Objective

Creating a predictive model using linear regression to analyse car sales and identification of
various key factors in predicting car sales

1.1.3. Scope of the Project

Scope of this project is global. It can be utilized by any business enterprise and firm or even an
individual interested in the related field.

1.2. Related Previous Work

 Market Sales Prediction using regression by Gagan Bhatia, using R

https://www.kaggle.com/gagandeep16/car-sales-prediction-using-regression
This is a car sales data set which is taken from Analytixlabs. This is a continuous data set which include
several predictors. From this data set they have to predict car sales by using machine learning
Techniques.

 Linear Regression Model applied in Used Vehicle Market by Mai Nguyen

https://towardsdatascience.com/car-selection-and-sales-day-prediction-8c4a474f9dca
This analysis is conducted to support purchasing managers at Drivetime for their selection process. The
main objective is to increase the proportion of vehicles sold within 90 days as well as to reduce the risk
of overage and auction vehicles, as well as maximize the profit for the company.

 Linear Regression Project by Benjamin Jones

https://www.kaggle.com/bdmj12/linear-regression-portfolio-project/data#Linear-Regression-
Project

Using Linear Regression to advise a (fake) Ecommerce company based in New York City that sells
clothing online but they also have in-store style and clothing advice sessions. Customers come in to the
store, have sessions/meetings with a personal stylist, then they can go home and order either on a
mobile app or website for the clothes they want.The company is trying to decide whether to focus their
efforts on their mobile app experience or their website.

1.3 Software and Hardware requirements

1.3.1 Hardware Requirements

 1 GB RAM
 1 GB hard free drive space

1.3.2 Software Requirements

 Jupyter Notebooks
 Pandas module
 Numpy Module
 MS Excel
 Web Browser: Microsoft Internet Explorer, Mozilla, Google Chrome or later
 Operating System: Windows XP / Windows7/ Windows Vista

1.4 Proposed Method

Model is created in 3 Stages:

 Part I : Pre Modeling

 Part II : Implementing Model (Modeling)
 Part III : Post Modeling

1.4.1 PRE MODELING

1. Understanding objective and deriving the Business problem
2. Validate the Problem
3. Convert Business Problem to Stats problem by defining possible X and Y, type of stat problem
4. Understand the data is required
5. Get the different data and create Data Audit Report at File Level

1.4.2 IMPLEMENTING MODEL

 1. Convert Business Problem to Stat Problem and based on the Stat problem at hand, select the
algorithm
2. Replace special characters with right/meaningful values
3. Perform Missing Value Treatment
4. Perform Outlier Capping
5. The data has to be split in train and test where on train data-set the model is built while on the test
the stability of the model is tested. The data is be split in 70-30 or 50-50 generally.
6. Coming up with a mathematical equation that can be used for predictions

1.4.3 POST MODELING

1. Come up with predictions.
2. Perform certain steps such as comparing the models performance on train v/s test.
3. Model object is to be exported for future use / implementation / productization.
4. Creating a Tool (Software, GUI etc.) where the model works at the backend.

1.5 Deliverables
Construction of a predictive model using Regression Analysis on the basis of Identified predictive
variables and validations of developed model for Business firms so that they can analyse the market
situation.

1.6 Stakeholders
Various Business firms and Business institutions and market analysts

1.7 Gantt Chart

1.8 References
Car Sales Prediction Using Machine Learning Algorithmns- Madhuvanthi.K, Nallakaruppan.M.K,
Senthilkumar N C, Siva Rama Krishnan S

Dataveda, book by Mr. Archish Rai Kapil