Datathon

It contains the solution to a competition from Kaggle.
(To run the codes successfully, you'll have to download the datasets from Kaggle)

Overview

In this datathon, we had access to a dataset containing information about various restaurants located in Bengaluru, collected from the popular restaurant and food delivery platform, Zomato. The primary task was to create a predictive model that estimates the cost associated with dining at these restaurants. Zomato defines this cost as the average expense required for two people to have a meal at a particular restaurant.

Dataset

The dataset provided included information on restaurants, such as their name, location, rating, type of cuisine, cost for two, and more. Understanding and manipulating this data was crucial in building an effective predictive model.

Technologies Used

• Python
• Jupyter Notebooks
• Data manipulation and analysis libraries (e.g., Pandas, NumPy)
• Data visualization libraries (e.g., Matplotlib, Seaborn)
• Machine learning libraries (e.g., Scikit-Learn, XGBoost)

Algorithm/Model Technologies Used

Extreme Gradient Boosting (XGBR)

Evaluation

The Root Mean Square Error (RMSE) is calculated as: $$RMSE = \sqrt{\frac{1}{n} \sum_{i=1}^{n} (y_i - \hat{y}_i)^2}$$

The RMSE measures the square root of the average squared differences between the actual and predicted values.