The work includes:
The theory of regression (linear vs logistic),
Assumptions of logistic regression,
Advantages and disadvantages,
Real-life applications (medicine, marketing, finance, fraud detection),
A case study with Titanic dataset to test the method.
Unlike linear regression (which predicts numbers like prices or scores), logistic regression predicts categories. For example:
Will a student pass or fail?
Will an email be spam or not?
Will a passenger survive or not?
It uses the Sigmoid function to give probabilities between 0 and 1. If the probability is greater than 0.5, the model predicts “Yes (1).” If lower, it predicts “No (0).”
The Titanic dataset is one of the most famous datasets in machine learning. It includes information such as:
Passenger class (1st, 2nd, 3rd),
Age, sex, ticket fare, cabin, port of embarkation,
Whether the passenger survived or not.
Steps in the project:
Data Review – looking at the dataset and its features.
Preprocessing – handling missing values.
Model Training – applying logistic regression.
Evaluation – checking the accuracy of the predictions.
The model could predict survival chances based on features like gender and class, which were among the most important factors.
Logistic regression is simple to apply and gives clear results.
It works well for binary outcomes.
However, it has some limits: it cannot predict continuous values, and small datasets can cause weak results.
The Titanic case showed that logistic regression is effective, but combining it with methods like decision trees can give even better accuracy.