Kavikulguru Institute of Technology and

Science,Ramtek
Project Preliminary Investigation Report

Name of Department:
Computer Technology

Name of Project Guide:

Mr. Bhushan A. Deshpande

Name of Project Co - Guide (if any):

Students Details :

Roll Name of Student Email ID Mobile No.

No.
01 Saurabh Ghute saurabhghute123@gmail.com 7350963234

02 Shreyash Kumbhare shreyashkumbhare483@gmail. 7387116166

com
03 Yeshwari Bhure yeshwaribhure49@gmail.com 9665591938

Ravikant Chandewar ravichandewar05@gmail.com 8530259527

04

Title of the Project:

Predictive Model for Retail Sales using Machine Learning
 1

Area of Project Work:

Machine Learning

Problem Statement:
To design sales predictive model using linear regression and random forest model
with the help of machine learning techniques and algorithms to forecast future
sales based on historical data, trends, and various influencing factors.

Prior Art (Patent Search):

Patent Title of Patent Existing Solutions
Applicati (Abstract of Patent)
on No.
US 9.202,227 B2 SALES PREDICTION This study explores the development and
SYSTEMS AND implementation of various sales prediction systems
METHODS and methods, focusing on their accuracy,
efficiency, and practical application in real-world
scenarios. This study investigate a range of machine
learning models and statistical techniques,
including Linear Regression, Decision Trees,
Random Forest, Gradient Boosting, Support Vector
Machines, ARIMA, SARIMA, and Long Short-
Term Memory (LSTM) networks. These models
are applied to a comprehensive dataset comprising
historical sales data, promotional activities,
seasonal effects, and economic indicators from a
retail company.
US PREDICTIVE AND This study explores the development and
2014/0067470 PROFILE LEARNING implementation of advanced analytics systems that
A1 SALES AUTOMATION combine predictive modeling and profile learning
ANALYTICS SYSTEM to automate and enhance sales processes. it utilizes
AND METHOD clustering techniques like K-Means and DBSCAN
(Density-Based Spatial Clustering of Applications
with Noise) for customer segmentation and profile
learning, enabling more personalized marketing
strategies.By apply these methods to a rich dataset
from a retail company, which includes historical
sales data, customer demographics, purchase
behaviors, promotional activities, seasonal effects,
and economic indicators.
 2

US 7.689,456 B2 SYSTEM FOR Accurately predicting sales lift and profit is critical
PREDICTING SALES for strategic decision-making in product
LIFT AND PROFIT OF A management and marketing. This study introduces
PRODUCT BASED ON a novel system designed to predict the sales lift and
HISTORICAL SALES profit of a product using historical sales
INFORMATION information. Leveraging advanced machine
learning algorithms and cloud computing
infrastructure, the system aims to provide precise
forecasts that can guide pricing strategies,
promotional activities, and inventory management.
WO FORECASTING SYSTEM The forecasting system's performance is evaluated
2023/215538 A1 USING MACHINE using metrics such as Mean Absolute Error (MAE),
LEARNING AND Mean Squared Error (MSE), and Root Mean
ENSEMBLE METHOD Squared Error (RMSE). The study also examines
the computational efficiency and scalability of the
proposed methods to ensure their practicality in
real-world applications. The dataset used for this
study comprises historical sales data from a retail
company, along with features such as promotional
activities, seasonal effects, and economic
indicators.
 3

Literature Review:

Title of Paper Details of Literature Identified for

Publication with Project
Date and Year
Predictive Analysis of February 2023 In this paper operation performed predictive
Retail Sales Forecasting analysis of retail sales of Citadel POS
using Machine Learning dataset, using different machine learning
Techniques techniques. Implemented different
regression (Linear regression, Random
Forest Regression, Gradient Boosting
Regression) and time series models (ARIMA
LSTM), models for sale forecasting, and
provided detailed predictive analysis and
evaluation. The dataset used in this research
work is obtained from Citadel POS (Point Of
Sale) from 2013 to 2018 that is a cloud base
application and facilitates retail store to
carryout transactions, manage inventories,
customers, vendors, view reports, manage
sales, and tender data locally
Benchmarking of January 2019 In this research, we employ a diverse set of
Regression and Time machine learning models and time series
Series Analysis analysis techniques, including Linear
Techniques for Sales Regression, Decision Trees, Random Forest,
Forecasting Support Vector Regression (SVR), Gradient
Boosting, SARIMA (Seasonal ARIMA),
networks. The dataset used for this study
comprises historical sales data from a retail
company, including features such as past
sales, promotional activities, seasonal
effects, and economic indicators.
Forecasting seasonals June 2004 Forecasting seasonal patterns and trends is
and trends by crucial for effective business planning and
exponentially weighted resource allocation. This study focuses on
moving averages using Exponentially Weighted Moving
Averages (EWMA) to forecast seasonals and
trends in time series data. EWMA is a
powerful tool for smoothing time series data
and making short-term predictions by
assigning exponentially decreasing weights
to past observations. By analyzing historical
data, we aim to provide accurate and reliable
forecasts of future trends.
 4

Drought forecasting in October 2016 This study advances drought modelling

eastern Australia using using multivariate adaptive regression
multivariate adaptive splines (MARS), least square sup port
regression spline, least vector machine (LSSVM), and M5Tree
square support vector models by forecasting SPI. . MARS model
machine and M5Tree incorporated rainfall as mandatory predictor
model with month (periodicity), Southern
Oscillation Index, Pacific Decadal
Oscillation Index and Indian Ocean Dipole,
ENSO Modoki and Nino 3.0, 3.4 and 4.0
data added gradually.
Market Basket Analysis: July 2016 Market Basket Analysis(MBA) also known
Identify the changing as association rule learning or affinity
trends of market data analysis, is a data mining technique that can
using association rule be used in various fields, such as marketing,
mining bioinformatics, education field, nuclear
science etc. The main aim of MBA in
marketing is to provide the information to
the retailer to understand the purchase
behavior of the buyer, which can help the
retailer in correct decision making.

Current Limitations
• Incomplete Data: Inconsistent or missing data can significantly impact model accuracy.
Retail data often have gaps due to incomplete transactions or system errors.
• Market Competition: Competitors' actions, such as pricing strategies or marketing
campaigns, can affect sales in ways that are difficult to predict.

Proposed Solution

• Developing accurate sales prediction model to avoid over forecasting and under forecasting
by using machine learning technique and algorithms.

• Appling linear regression, LSTM, ARIMA, and random forest model sales data prediction
accuracy will be increases.
 5

Objectives and Scope of Work

• To develop a sales predictive model using machine learning that can accurately predict
retail sales for over-forecasting and under-forecasting method.

• To develop accurate and robust forecasting model by implementing the random forest
model and linear regression methods for future sales prediction

Feasibility Assessment:

I. Expected Outcomes of the Project

- Accurate Sales Forecasting

- Enhanced Decision-Making

-Increased Revenue

II. Innovation Potential

- Convolutional Neural Networks (CNNs): For image-based data (e.g., product

images) to understand visual attributes influencing sales.

- Reinforcement Learning: To optimize pricing, promotions, and inventory levels

through trial-and-error learning.

III. Task Involved

- Project Planning and Management

- Data Collection and Integration

 6

IV. Expertise Required

1. Inhouse Expertise
-Guidance of project coordinator

2. External Expertise
-Not required

V. Facilities Required

1. Inhouse Facilities
- Computer with necessary software and hardware.
-Access to industrial equipment dataset.

2. External Facilities
-Not required
 7

Milestones and Time Plan

Task JUL AUG SEP

2024 2024 2024

Conceptual Design
✓
Design
Detailed design
✓
Design Modifications
✓
Final Design
✓
Procurement (If any)
✓
Develop Prototyping
✓
Modifications
✓
Testing and Validation
✓
Final Modifications
✓
Deliver
IPR / patent draft
✓
Thesis and Poster
✓

Name and Signature of Project Guide Signature of HOD

8