Inspiration
Prostate cancer is a major health issue worldwide, highlighting the need for early detection based on specific symptoms.
What It Does
This machine learning model predicts the early detection of prostate cancer using a dataset from Kaggle.
How We Built It
We developed the model using Jupyter Notebook and Python, utilizing scikit-learn for machine learning.
Challenges We Faced
One of the main challenges was achieving higher accuracy, with the maximum accuracy currently at 91%.
Accomplishments We're Proud Of
We explored multiple models and approaches to improve performance. (An accuracy comparison table will be provided below.)
What We Learned
We gained insights into various techniques and methodologies for building an effective ML model.
What's Next for the Prostate Cancer Prediction ML Model
Our next goal is to further improve the model’s accuracy and enhance its effectiveness for early detection.
Accuracy Table
| Model | Accuracy Score |
|---|---|
| Random Forest Classifier | |
| Random Forest Classifier | 0.84947 |
| Random Forest Feature Importance | 0.84947 |
| Random Forest Mutual Information | 0.84840 |
| Random Forest Chi Feature | 0.84935 |
| Random Forest SMOTE | 0.91066 |
| Random Forest SMOTE PCA | 0.61190 |
| Decision Tree Classifier | |
| Decision Tree Classifier | 0.72280 |
| Decision Tree Feature Importance | 0.72614 |
| Decision Tree Mutual Information | 0.72650 |
| Decision Tree Chi Feature | 0.72268 |
| Decision Tree SMOTE | 0.83311 |
| Decision Tree SMOTE PCA | 0.60165 |
| Logistic Regression Classifier | |
| Logistic Regression Classifier | 0.84947 |
| Logistic Regression Feature Importance | 0.84947 |
| Logistic Regression Mutual Information | 0.84947 |
| Logistic Regression Chi Feature | 0.84947 |
| Logistic Regression SMOTE | 0.51982 |
| Logistic Regression SMOTE PCA | 0.50557 |
Built With
- jupyter
- notebook
- python
- scikit-learn
Log in or sign up for Devpost to join the conversation.