Welcome to the "Kernelizing Cells" project! This repository utilizes Support Vector Machines (SVM) with a linear kernel to detect cancer in human cell samples. The dataset, sourced from the UCI Machine Learning Repository, includes records with various cell characteristics. The analysis involves training an SVM model, evaluating its performance, and extracting meaningful insights from the dataset.
- Python
- Jupyter Notebooks
- Scikit-learn
- Matplotlib
- Pandas
- NumPy
- scikit-learn
- matplotlib
- pandas
- numpy
- Implementation of SVM with a linear kernel for binary cancer classification.
- In-depth exploration of dataset characteristics and feature importance.
- Utilization of key evaluation metrics, including F1-score and Jaccard score.
- Machine Learning: Employing SVM for accurate cancer detection.
- Data Analysis: Uncovering hidden patterns through statistical analysis.
- Model Evaluation: Leveraging F1-score, Jaccard score, and other metrics for robust assessment.
- Attained a high F1-score (0.9639) and Jaccard score (0.9444) showcasing the model's efficacy.
- Unveiled insights into feature importance and intrinsic dataset patterns.
- This repository offers a comprehensive exploration of cancer detection using SVM.
- Provides detailed insights into the model, evaluation metrics, and underlying dataset patterns.
- A valuable resource for those interested in SVM, binary classification, or cancer diagnostics.
Feel free to explore the Jupyter Notebooks for a detailed walkthrough of the project. Contribute, share, and adapt this code to elevate your own analyses.
For further details, refer to the documentation and notebooks within the repository. Let's Kernelize Cells for a brighter future in cancer detection!