Skip to content

yaseminates/Optimized-ML-Model-Data-Analysis-for-Business

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

22 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Optimized-ML-Model-Data-Analysis-for-Business

Group Name: Bayes-ically_Unstoppable

2025-03-30

Introduction

This is a Group Project for the Data Analysis for Business class of LUISS University’s Management and Computer Science degree. a.y. 2024/2025 - Spring. The project was awarded a 30/30.

  • 296881 Guia Ludovica Basso
  • 297061 Alessio Giannotti
  • 304011 Yasemin Ateş

View the output here: https://yaseminates.github.io/Optimized-ML-Model-Data-Analysis-for-Business/

This project analyzes the "Hitters" dataset to predict salary levels by converting continuous salaries into categorical groups and modeling them using multinomial logistic regression. The analysis includes data cleaning, exploratory data analysis, feature evaluation, and model assessment to identify the most relevant predictors of salary.

The dataset contains performance and salary information for Major League Baseball (MLB) players from the 1986 and 1987 seasons. Each row represents a player, with variables capturing both seasonal and career statistics.

Conclusion

The aim of this analysis was to predict baseball player’s salaries as “Low”, “Medium”, “High” or “Very High” as determined by clustering the collected salary values.

We can conclude that the more seasoned and experienced a player is (in relation to the Years feature), and the better they perform (in relation to the performance metrics like Hits, HmRun, etc.) the higher their salary will be. We further analyzed this ny creating a success score and saw the positive relation proven once again. The league or division of the player seems not to have too large an effect on the salary, especially compared to their performance.

In further data gathering practices or analysis, more entries collected over a longer period of time will surely help refine these models.

About

ML model optimization for classification on highly skewed data. View the HTML output of the R code below.

Topics

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors