Predicting Pizza Costs: An Evaluation of Random Forest and TPOT AutoML

  • Abhishek Singh Associate Professor, Dept. of CSE., BGIEM, Jabalpur, India
  • Zohaib Hasan Associate Professor, Dept. of CSE., BGIEM, Jabalpur, India
  • Nirdesh Jai Associate Professor, Dept. of CSE., BGIEM, Jabalpur, India
Keywords: Machine learning, Random Forest, TPOT AutoML, Data Augmentation, Food industry pricing

Abstract

In recent years, pizza has become more and more popular in India. Pizza has been available in the US for a while, but in recent years, its craze has really taken off. Important pizza-related businesses are thus seeing great investment and development opportunities in this area. Pizza is quickly rising in popularity as the preferred food for many Indian residents, and the number of restaurants serving it is expanding. In order to anticipate pizza prices, this study assesses how effective the two machine learning techniques Random Forest and TPOT (Tree-based Pipeline Optimization Tool) AutoML are. For model training and assessment, a dataset comprising several pizza variables, like diameter, toppings, and extra ingredients, was utilized. Metrics like mean squared error and R-squared were used to evaluate the prediction accuracy of both models after they had been trained and tested according to normal protocols. The findings suggest that while TPOT AutoML performs somewhat better in some circumstances, Random Forest and TPOT AutoML both exhibit encouraging performance in forecasting pizza costs. These results demonstrate how well machine learning methods work to forecast intricate pricing schemes in the food sector.
Published
2024-03-25