You are on page 1of 18

Recommendation of Indian Recipes

based on Ingredients

Presented By:
Deepali Matkar (Roll No:52)
Sonal Nalbalwar (Roll No:56)
Tejaswini Mane (Roll No:61)
Motivation

• Thinking of what to cook is a difficult problem . To


attract children liking, parent need to exchange the
menu everyday, they also need to consider the
nutrition that their children taken. Besides that
some people will forgot to buy ingredients to stock
in their kitchen. By considering all this facts the we
got decided the idea of implementing recipe
recommendation system
Aim and Objectives
Aim:
To develop a recommender system that carries out the
process of suggesting new cuisines to the user.
Objectives:
1.To develop a system that recommend recipe to
users from the available ingredient.
2. Enable the users upload the recipe to enlarge the
database
Problem statement
• Thinking of what to cook is a difficult problem, when
we have some ingredients
S Title Author Advantages Disadvantages Methodology
r.
N
o

1 A scalable ZhengXia User Mobility, Hybrid A hybrid system with


recipe n Li, Device Mobility recommendation pipelined
recommen Jinlong and Wireless system recommender approach
dation Hu*, Access to improve the recipe
system for Jiazhao
recommended
mobile Shen,
application Yong Xu effectiveness

2 NutRec: Elizabeth 1.Providing Influence our food 1.NutRec: Recommend


Nutrition Gorbonos recipe based on preferences among 2.Pseudo-recipe
Oriented ∗, Yang user nutrition which are construction
Online Liu†, 2.generate geographical 3. Ingredient predictors
Recipe Chınh T.
healthy and cultural 4. Amounts predictor
Recommen Hoang
der recommendation affiliation
over an online
non-healthy
dataset
sr Title Author Advantages Disadvantages Methodology
n
o

3 optimization Isura Provided the Class imbalance Data mining,


Framework Nirmal#1, ingredients when classifying Personalization
for Flavour Amith suggestions the recipes into recommendation
and Caldera#2, which are cuisines. approach,
Nutrition Roshan
optimized in Generalized
Balanced Dela
Recipe: A Bandara both flavour and recommendation
Data Driven nutrient approach
Approach

4 A Recipe Lin-rong 1.User can find Algorithms used Term Frequency Inverse
Recommend GUO, Shi- different cuisines did not study the Document Frequency to
ation System zhong of similar flavour quantity of calculate ingrediant
Based on YUAN*, 2. High ingredients in the preferences, Tidal-Trust
Regional Xue-hui
frequency recipe algorithm
Flavor MAOand
Similarity. Yi-ning ingredient
GU inquiry of
various cuisines
sr Title Author Advantages Disadvantages Methodology
n
o

5 Cuisine S. Kalajdziski1, 1.User can Misclassification errors support vector


classificati G. Radevski2, identify occur when a recipe machines, neural
on using I.Ivanoska3, K. cuisine just by that belongs to a less networks for
recipe’s Trivodaliev4 and ingredients frequent class is recipes
ingredients B. Risteska
mistakenly classified as classification
Stojkoska5
a recipe from one of the
most frequent classes
Existing system
• Recipe Recommendation System for Indian cuisines
is a system that learns from the past preferences of a
user’s preferred dishes to recommend him/her new,
untested cuisines. The basis of recommendations is
the ingredients in the recipes already liked by the
user.
• Disadvantages:
1.Web scrapping
2.Waste of time and money.
Existing system Vs Proposed system
Existing system Proposed system

Recipe Recommendation System for The flow by which the recommender system
Indian cuisines is a system that learns carries out the process of suggesting new
from the past preferences of a user’s cuisines to the user. For each user, the
preferred dishes to recommend him/her system stores the dishes liked by him/her,
new, untested cuisines. In this system we and then tries to match their ingredients with
search for the recipe then we get the list the ones in its database. Proposed system
of ingredients. show the nutrition value of recipe and
suggest user to try the same recipe with
healthier ingredients. It avoids the web
scrapping.
Algorithm
• Collaborative filtering : collaborative filtering (CF)
is making recommendations based on other
comparable customers' profile. It is useful into
exceptionally assorted arrangements of things. The
resemblances between different things in the dataset
are figured by using one of the different closeness
measures, and a while later, these comparable
qualities are used to foresee examinations for
customer thing sets not present in the dataset.
System Architecture
Flowchart
Use case Diagram
Sequence Diagram
Activity Diagram
Class Diagram
Software & hardware Requirements
• HARDWARE REQUIREMENTS:

• System : Intel I3 Processor and above.


• Hard Disk : 20 GB.
• Ram : 3 GB and above.

• SOFTWARE REQUIREMENTS:

• Operating system : Windows 7 and above, Android.


• Coding Language : Java 1.8
• IDE : Eclipse
• Database : MySQL
References
[1] Cuisine classification using recipe’s ingredients S. Kalajdziski1 , G.
Radevski2 , I.Ivanoska3 , K. Trivodaliev4 and B. Risteska
Stojkoska5 Ss. Cyril and Methodius University, Faculty of
Computer Science and Engineering, Skopje, Macedoni.

[2] Optimization Framework for Flavour and Nutrition Balanced


Recipe: A Data Driven Approach Isura Nirmal#1, Amith
Caldera#2, Roshan Dela Bandara*3 # University of Colombo
School of Computing, Colombo, Sri Lanka

[3] A scalable recipe recommendation system for mobile application


Zheng Xian Li, Jinlong Hu*, Jiazhao Shen, Yong Xu 2016.

You might also like