A machine Learning app for snack suggestion | The goal for this project is to build a system that allows you to identify and then recommend, recipes you're likely to enjoy.
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
ML
database
epicurious-recipes-with-rating-and-nutrition
images
models
static
templates
.DS_Store
README.md
app.py
rec_table_creator.py

README.md

Snack Search

Author: Rishab Sharma - MAIT, GGSIP University ,New Delhi - 110086

A machine Learning Driven app for snack suggestion

One of the main uses of computers is to help us solve problems quickly and effectively. And a problem we often run into is figuring out what to eat, or what to make. This problem is solvable using data and recommendation engines.

Recommendation engines work on two levels. The first level is on the personal level. Let's say you create a dataset of foods and rank how much you enjoy or dislike them, 1-10. Given an unseen food and its set of features (such as the inclusion of ingredients, or perhaps the percentage of that meal the ingredient takes up). A machine learning algorithm figure out if and how much you'd like it. The other way recommendation engines work is on the group level. A machine learning algorithm should be able to recommend new foods to you, given a set of people who share your similar food preferences.

The goal for this project is to build a system that allows you to identify and then recommend, recipes you're likely to enjoy.

I have used Flask Microframework to serve my app.

Home Screen

home

Enter your Habits

Multi - Algorithm Recommender

Recommmender

Support Vector Machine - Output

Random Forest - Output

Decision Tree - Output

Neural Network - Output

Tech Stack

Languages

  1. Python
  2. HTML
  3. CSS
  4. Javascript
  5. Jquery

Libraries

  1. ScikitLearn
  2. Keras
  3. Tensorflow
  4. Flask
  5. Python Numpy, OS , Sys , Matplotlib

Database

SQLite3

Dataset

Kaggle

Thank You , Fork If you Find it Useful