You are on page 1of 19

DR.

AKHILESH DAS GUPTA INSTITUTE OF TECHNOLOGY AND


MANAGEMENT

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING

Snac a Treat with Recommendation


System

Under the guidance of


Dr. Shivangi

Presented by :
Shrey [03396202718]
Chirag[01096202718]
Pankaj[02796202718]

Date:03-01-2022
Department of Computer Science &
Engineering
VISION
To produce socially responsible technocrats, researchers , and
entrepreneurs in the field of computer science and engineering.

MISSION
 
M1 -To provide quality education in the field of computer science and
engineering with emphasis on research and innovations.
M2 - To inculcate professional behaviour , strong ethical values, and
leadership skills.
M3 - To provide a platform for promoting entrepreneurship and
multidisciplinary activities.

2
Table of contents
• Objective
• Introduction
• Methodology with Block Diagram
• Nature of Project: Software
• Final Application
• Name of three SCI/Scopus indexed journals or
IEEE conference in one of which work will be
published
3
Objective
• The main objective of the project is to provide a
platform which connects customers, restaurant partners
and delivery partners, serving their multiple needs with
recommendation system.

• Our aim is to suggest products to the customers. The


products can be recommended based on the top overall
sellers on a site, based on the demographics of the
customer, or based on an analysis of the past buying
behavior of the customer as a prediction for future buying
behavior.
4
Introduction
• Snac A Treat is an ecommerce simulated website. We provide
restaurant partners with industry-specific marketing tools which
enable them to engage and acquire customers to grow their
business while also providing a reliable and efficient last mile
delivery service.

• Customers use our platform to search, read and write customer


generated reviews and, order food delivery, and make payments.

• The user can also add the particular food as suggested by the
recommendation system used in the project.
• Special color blind mode
5
• Admin Dashboard to accept and decline the order.
• Filtering system to filter the data according to price
range, rating and availability.
• On the other hand, We also operate a one-stop
procurement solution. We also provide our delivery
partners with transparent and flexible earning
opportunities.
Recommendation System
LINK :
https://food-recommendation-pk.herokuapp.com/
● This is a machine learning web app, when executed, the client gets to choose a food
item based on his own liking, and we recommend 5 more such similar foods through
our recommendation system.
● Our system utilizes item-based collaborative filtering to implement the same.The rating
scale can be set by the user.
● Recommendation System using Nearest Neighbour.
● Library Used :
○ Pickle : Used here to save ml model them to be able to make new predictions at a later time or to
load an already prepared model.
○ Streamlite : An open-source python framework for building web apps. We can instantly develop
web apps and deploy them easily using Streamlit.
○ Numpy : Used for complex mathematical operations. NumPy overcomes slower executions with the
use of multi-dimensional array objects.
○ Pandas : Used for data.
○ sklearn.neighbors : For the simple task of finding the nearest neighbors between two sets of data,
the unsupervised algorithms within sklearn.neighbors can be used.
7
Collaborative Filtering
Techniques related to memory based collaborative filtering
To generate prediction Memory-based CF algorithms uses the total or a some part of
database of the user-item. Here every user with similar interests is part of a similar group
of people. By identifying the neighbors of a new user or currently active user, it can
produces a anticipation of preferences on new items for him or her.
• k nearest neighbors The most extensively used algorithm for collaborative filtering is the k Nearest
Neighbors (kNN). There are two types of k Nearest Neighbor algorithms:
1. User based Nearest Neighbor.
2. Item based Nearest Neighbor.
• We have used Item Based Nearest Neighbor:
The item-based approach investigates the set of items rated by target user and calculates their
similarity with the target item i and then chooses k most similar items 𝑖1 , 𝑖2 , … 𝑖𝑘 . Their representing
similarities 𝑠𝑖1 , 𝑠𝑖2 , … 𝑠𝑖𝑘 are also computed at the same time. Similarity computation and the
prediction generation are two important factors which make item-based recommendation more
powerful. For similarity computation basically different types of similarity measures are used and
weighted sum and regression used for prediction computation.

8
Methodology/Block Diagram
System Model:

Admin
9
Flow of Project:

Home

Item=0
Select Product My Cart Item Continue
Menu
Shopping
Item>0
Select Proceed to No
Login Sign In
Category checkout
Yes

View Recommended Payment


Product Information Correct
Product

No Yes
Review
Add to Cart Order

Order
Confirmation Check Out

10
Nature of Project- Software
Technologies used-
1)Frontend - HTML, CSS, JavaScript, ReactJS,
Tailwind CSS, Postcss, Redux.
2)Backend- Node, Express, AJAX, MongoDB,
Cloudinary, Payment Gateway( Pay Stack)
3)Deployment- Nginix based AWS

11
Final Application
● Home Page

12
●Menu Page

13
●Recommendation Page

14
●Cart Page

15
●Payment Page

16
• Admin Page

17
• Payment receipt

18
Research Paper
• REFERENCES :
1).KirtiBhandge, TejasShinde, Dheeraj Ingale, Neeraj Solanki, ReshmaTotare,”A
Proposed System for Touchpad Based Food Ordering System Using Android
Application”, International Journal of Advanced Research in Computer Science
Technology (IJARCST 2020).
2). VarshaChavan, PriyaJadhav,Snehal Kharade,PriyankaTeli, ”Implementing
Customizable Online Food Ordering System Using Web Based Application”,
International Journal of Innovative Science, Engineering Technology(IJISET) 2015.
3). J. B. Schafer, D. Frankowski, J. Herlocker, S. Sen,“Collaborative filtering
recommender systems”, in: P. Brusilovsky, A. Kobsa, W. Nejdl (Eds.), The Adaptive
Web, 2007, pp. 291–324 .

19

You might also like