The main goal of this project is to apply suitable machine learning algorithms learned in class to solve a specific data science problem, in this case, models capable of identifying benign or malignant masses in mammography’s. In this report, we try to obtain the best result of the accuracy of the implemented models.
The project was developed using the conda environment, so there is no file requirements.txt, but the libraries used were:
- numpy
- matplotlib
- pandas
- seaborn
- sklearn
The dataset used in this project is the Mammographic Mass Data Set, which can be found at the following link and is available here in the repository too.
The machine learning process was divided into 4 steps:
- Data Preprocessing
- Data Visualization
- Model Training
- Model Evaluation
All the code can be found here in the Jupiter Notebook.
The report of the project can be found here. The report was made in LaTeX and the source code can be found here as well.