Topic: An analysis of American Airlines Traffic, Capacity and Revenue data by Operating Region between 1995 - 2020.
- To determine which of the operating regions has been profitable for America Airlines.
- To establish the relationship between revenue and operating expenses.
The project uses data with 83000 rows in predicting the store sales for various company products. The naive model proved to be the most suitable predictor compared to the ARIMA model.
This is a simple project using a simple data of the list of music various categories of adults and gender (male and female) to predict the type of music a certain person may like based on age and gender.
The Model has an accuracy scope ranging between 75% to 100%. The model was tested and confirmed that it provides accurate predictions. In this model, i use JobLib package to create the model and save in my machine to avoid retraining of the model and maintain its persistence.
This project uses the titanic data and Python machine learning to predict the survival of different categories of people based on fare, age, sex and passenger class. The machine learning model has an accuracy score of 95%. i. Funny findings from the model prediction results: Fare was a major determinant for the survival rate followed by gender and Pclass. ii. Men who paid less fare did not survive compared to women with the same age and who paid the same fare.
This project focuses on analyzing the most important factors that contributes to employee turnover. Th project focuses on examining factors such as salary, satisfaction level, promotion in the past 5 years, work accidents, and time spent in a company. I created a machine learning model that uses 15000 rows of data to identify the factors in various departments. The model has an accuracy of ~80%, but the results tested so far are 100% accurate. Within the Python notebook, i have provided detailed insights on the ML model test. This project was inspired from https://www.kaggle.com/code/jacksonchou/hr-analytics/report.