https://github.com/reshmaaiman/fifa

FIFA20
https://github.com/reshmaaiman/fifa

data data-science data-visualization dataanalysisusingpython github jupyter-notebook matplotlib numpy pandas python seaborn-python

Last synced: 15 days ago
JSON representation

FIFA20

• Host: GitHub
• URL: https://github.com/reshmaaiman/fifa
• Owner: ReshmaAiman
• Created: 2025-02-04T15:20:26.000Z (17 days ago)
• Default Branch: main
• Last Pushed: 2025-02-04T15:44:35.000Z (17 days ago)
• Last Synced: 2025-02-04T16:33:43.754Z (17 days ago)
• Topics: data, data-science, data-visualization, dataanalysisusingpython, github, jupyter-notebook, matplotlib, numpy, pandas, python, seaborn-python
• Language: Jupyter Notebook
• Homepage:
• Size: 17.9 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# FIFA

### The objective is to explore football skills and cluster football players based on their attributes so that we can easily understand who is the best player and with the FIFA20 dataset we need to cluster the player by their skills and group them to appropriate group.

### Python libraries used - numpy, pandas, matplot, seaborn, scikit learn, sweetviz.

### The data had lot of dupicates and unneccesary data, followed data pre processing steps to obtain clean data for better results.

### Performed Bivariate analysis to understand how each attribute is dependant on the other.

### After succesful analysis, determined that most valuable player in FIFA20 is Neymar Jr followed by Messi & K.Mbappe.

### When we take top 5 valuable players in FIFA20, two of them are playing for the Paris Saint-Germain.

### Messi is getting hifher wage followed by Eden Hazard and Christiano Ronaldo.

### The order of tp 10 Couties with the most players in FIFA20 is - England, Germany, Spain, France, Argentina, Brazil, Italy, Columbia, Japan, Netherlands.

### Also determined from the data that left wingers are paid most.

### Tried with different models like K-Means, Hierarchical clustering, DBSCAN and choosed DBSCAN with the silhouette score being 0.79 for the data.