智慧物流>> PCA-Clustering-and-SVC-SVM-McDonalds-Menu>> 返回
项目作者： jneitman

项目描述 ：
PCA, Clustering, and SVC/SVM for items on the McDonald's Menu.
高级语言：
项目地址: git://github.com/jneitman/PCA-Clustering-and-SVC-SVM-McDonalds-Menu.git


Principle Component Analysis, Clustering, and Support Vector Classifiers/Support Vector Machines using Data from the McDonald’s Menu

This project was submitted as a final deliverable for a data mining class offered through the M.S. Data Science program at the University of Wisconsin - Eau Claire.

The project focuses on past and current items from the McDonald’s menu with a goal of calculating and visualizing principle components and clusters (k-means and hierarchical) in relation to nutritional data. Additionally, support vector classifiers and support vector machines where double cross-validated to obtain a model capable of predicting the category of future items based on its nutritional data. A support vector classifer (kernel = linear) resulted as the best performing model.

Figures and tables located in McDonalds_Menu_PCA_Clust_SV.md file.

Project Contents

  1. Loading the data and exploratory analysis
  2. Principle Component Analysis
  3. Hierarchical clustering
  4. K-mean clustering
  5. Cross-validation and double cross-validation of SVC and SVM
  6. Fitting data to best model and predicting categories of future items