Breast Cancer prediction using Ml models and Deep learning models like cnn and ann v2
Budget: ₹1,500 – ₹12,500 INR
Basically the data set is the breast Cancer data set. -breast-cancer-wisconsin (1).data
Development of a model to identify the breast cancer with high degree of precision.
Need end to end project with UI interface to show the below steps .
reference method
https://medium.com/analytics-vidhya/building-ml-model-to-predict-whether-the-cancer-is-benign-or-malignant-on-breast-cancer-wisconsin-d6cf8b47f49a
Result - need to compare ML models with CNN and ANN . and see if Deep learning model are better than ML models .
The work-flow will follow these steps:
1. Import Data
2. Analyze data structure and statistics a. Clean Data (handle with outliers and missing values)→ EDA
3. Create Visualizations to better understand data
4. Create new features / Feature selection - Identify the best features or combination of features
5. Test dimension reduction using PCA
6. Test, measure and tune different methods
• Test classifiers (e. g. XGBoot, Decision Trees, Support Vector Machine, Logistic Regression, Bayes, Also try CNN based approach, ensemble ..etc)
• Test results and select some of the bests algorisms
• Tune algorithms parameters
7. Deploy the model based on the optimal accuracy and other parameters .
8. Analyze results and make conclusion
ML Workflow:
1. EDA and feature selection (Data preprocessing, Data Munging/ Wrangling)
2. Data Modeling - Testing with multiple models and finalizing the right model
3. Validation, Testing and optimization
4. Deployment and presentation
Techniques: Classical Machine Learning, Deep Learning
Development of a model to identify the breast cancer with high degree of precision.
Need end to end project with UI interface to show the below steps .
reference method
https://medium.com/analytics-vidhya/building-ml-model-to-predict-whether-the-cancer-is-benign-or-malignant-on-breast-cancer-wisconsin-d6cf8b47f49a
Result - need to compare ML models with CNN and ANN . and see if Deep learning model are better than ML models .
The work-flow will follow these steps:
1. Import Data
2. Analyze data structure and statistics a. Clean Data (handle with outliers and missing values)→ EDA
3. Create Visualizations to better understand data
4. Create new features / Feature selection - Identify the best features or combination of features
5. Test dimension reduction using PCA
6. Test, measure and tune different methods
• Test classifiers (e. g. XGBoot, Decision Trees, Support Vector Machine, Logistic Regression, Bayes, Also try CNN based approach, ensemble ..etc)
• Test results and select some of the bests algorisms
• Tune algorithms parameters
7. Deploy the model based on the optimal accuracy and other parameters .
8. Analyze results and make conclusion
ML Workflow:
1. EDA and feature selection (Data preprocessing, Data Munging/ Wrangling)
2. Data Modeling - Testing with multiple models and finalizing the right model
3. Validation, Testing and optimization
4. Deployment and presentation
Techniques: Classical Machine Learning, Deep Learning