CreatePerfectRecipe

• Our CS 478 Team Project consists of gathering/formatting data and running several machine learning algortihms on data from recipe sites.

Objective

• Train machine learning algorithms to rate recipes based on:
  • Ingredients
  • Time
  • Nutritional Value
  • Etc

Phases

1. We use the Chrome Extension WebScraper to get the raw data from recipe sites such as AllRecipes.
2. We will then run some algorithms on the data to format it into a common format that can be run through a machine learning algorithm.
3. Finally we will train several machine learning algorithms (from tensorflow) on the data.
4. We will test the machine learning model on novel recipes from some other site and compare them to reviews and try some of the best/worst ones out.
5. We will compile the results and present them to the class.

Tools

• We will be programming/scripting in Python 3.6 using Anaconda
• We will be using Jupyter Lab and Jupyter Notebooks from Anaconda where it makes sense
• We will be using MongoDB for storing the data
• We will be using pymongo v 3.4 for interacting with MongoDB
• We will be using tensorflow for machine learning

Repo Organization

Organization will go as follows: Each folder will have a README.md with an explanation of what is in the folder and any subfolders

• python is the folder where all of our code will go (both jupyter notebooks and python scripts)
• data is the folder where all data goes (database and scrapers)