/Restaurant-Turnover-Prediction

Hackathon Machine Learning Project for the Post Graduate Program in Artificial Intelligence & Machine Learning: Business Applications (PGP-AIML) by McCombs School of Business at The University of Texas at Austin offered in collaboration with Great Learning.

Primary LanguageJupyter Notebook

Restaurant Turnover Project

Hackathon

Predict the Annual Turnover of a restaurant based on the variables provided in the data set.

Project Overview

Looking at this from the business perspective for a restaurant, more popularity may mean more visits to the joint increasing the annual turnover of the restaurants. For any restaurant to survive and do well, the annual turnover of the restaurants has to be substantial.

This problem takes a shot at predicting the annual turnover of a set of restaurants across India based on a set of variables given in the data set. This includes the data related to the restaurant such as location, opening date, cuisine type, themes etc. This also includes data pooled from different sources such as social media popularity index, Zomato ratings, etc. Lastly, it also adds a different flavour to the problem by looking at the Customer survey data as well as ratings provided by a mystery visitor data (audit done by a third party).

Objective

Data has been split into two groups and provided in the module:

  • Train dataset
  • Test dataset

The train dataset set is used to build your machine learning model. For the training dataset, we provide the Turnover of the restaurant (also known as the variable Turnover) for each participant.

The test dataset should be used to see how well your model performs on unseen data. For the test dataset, it is your job to predict the Turnover of the restaurant (Turnover) for each participant.

In Data Dictionary_(1).csv, the detail of all the variables used in the train and test datasets is given.