https://github.com/srosalino/clustering_on_starbucks_beverages

Grouping of drinks according to their nutritional values, making it easier to categorize them in a future catalog, increasing organization and facilitating the search depending on individual preferences
https://github.com/srosalino/clustering_on_starbucks_beverages

gaussian-mixture-models hierarchical-clustering k-means-clustering pam pca unsupervised-learning

Last synced: 4 days ago
JSON representation

Grouping of drinks according to their nutritional values, making it easier to categorize them in a future catalog, increasing organization and facilitating the search depending on individual preferences

• Host: GitHub
• URL: https://github.com/srosalino/clustering_on_starbucks_beverages
• Owner: srosalino
• Created: 2024-07-11T15:53:28.000Z (4 months ago)
• Default Branch: main
• Last Pushed: 2024-07-11T15:56:47.000Z (4 months ago)
• Last Synced: 2024-07-11T18:22:18.973Z (4 months ago)
• Topics: gaussian-mixture-models, hierarchical-clustering, k-means-clustering, pam, pca, unsupervised-learning
• Language: Jupyter Notebook
• Homepage:
• Size: 5.95 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
**Overview**

Starbucks is an American multinational company that presents itself as one of the largest coffee shop chains in the world. Created in 1971 by Jerry Baldwin, Zev Siegl and Gordon Bowker in Seattle with the opening of a simple coffee shop, the company currently has almost 30,000 stores around the world. Nowadays, food and nutrition are increasingly given high importance, mainly due to greater health concerns. Therefore, it is important to create menus that better guide customers when choosing the product to be consumed, categorizing them according to their nutritional compositions. Starbucks, not shying away from innovation in this sense, intends to make available to its customers a catalog of its various drinks taking into account their different nutritional characteristics. To this end, a detailed study was requested based on a set of data made available by the company itself. The objective of this study will be to group drinks according to their nutritional values, making it easier to categorize them in a future catalog available in physical or digital version (in your application), increasing organization and facilitating the search depending on individual preferences. The following report presents the various phases and techniques that were implemented in order to achieve the proposed objective. Therefore, we used dimensionality reduction techniques such as Principal Component Analysis (PCA) followed by Unsupervised Learning techniques, namely grouping observations, that is, building clusters. Within the application of Clustering, several techniques are used such as hierarchical clustering, partitioning techniques such as K-means and PAM (Partition around medoids) and probabilistic clustering models, such as GMM (Gaussian Mixture Model).

**Full Report**

Please access the file '*Relatório_GR11_MANS.pdf*' for a detailed explanation of the work developed and obtained results.