abhisheks008 · abhisheks008 · Jan 16, 2024 · Jan 15, 2024
diff --git a/Critically Ill Patients Analysis and Prediction/Dataset/README.md b/Critically Ill Patients Analysis and Prediction/Dataset/README.md
@@ -0,0 +1,11 @@
+# Critically Ill Patients Dataset 
+
+The Dataset used here is taken from the Kaggle database website. You can download the file from the link given here, [Critically Ill Patients Analysis and Prediction](https://www.kaggle.com/datasets/margaritakholostova/support-ii-dataset-with-critically-ill-patients)
+
+## About the dataset
+
+There are 2 types of datasets:
+
+- `columns_descriptions`: This dataset contains the description of all the features.
+- `support2`: This dataset contains 9105 entries with 47 different features.
+
diff --git a/Critically Ill Patients Analysis and Prediction/Images/age_plot.jpg b/Critically Ill Patients Analysis and Prediction/Images/age_plot.jpg
diff --git a/Critically Ill Patients Analysis and Prediction/Images/correlation_heatmap.jpg b/Critically Ill Patients Analysis and Prediction/Images/correlation_heatmap.jpg
diff --git a/Critically Ill Patients Analysis and Prediction/Images/diabetes_plot.jpg b/Critically Ill Patients Analysis and Prediction/Images/diabetes_plot.jpg
diff --git a/Critically Ill Patients Analysis and Prediction/Images/dnr_plot.jpg b/Critically Ill Patients Analysis and Prediction/Images/dnr_plot.jpg
diff --git a/Critically Ill Patients Analysis and Prediction/Images/edu_plot.jpg b/Critically Ill Patients Analysis and Prediction/Images/edu_plot.jpg
diff --git a/Critically Ill Patients Analysis and Prediction/Images/temp_plot.jpg b/Critically Ill Patients Analysis and Prediction/Images/temp_plot.jpg
diff --git a/Critically Ill Patients Analysis and Prediction/Images/urine_plot.jpg b/Critically Ill Patients Analysis and Prediction/Images/urine_plot.jpg
diff --git a/...ients Analysis and Prediction/Model/critically_ill_patients_analysis_and_prediction.ipynb b/...ients Analysis and Prediction/Model/critically_ill_patients_analysis_and_prediction.ipynb
diff --git a/Critically Ill Patients Analysis and Prediction/README.md b/Critically Ill Patients Analysis and Prediction/README.md
@@ -0,0 +1,86 @@
+<h1>Critically Ill Patients Analysis and Prediction</h1>
+
+**GOAL**
+
+To build a machine learning model for predicting the patient survival rate or probability.
+
+**DATASET**
+
+https://www.kaggle.com/datasets/margaritakholostova/support-ii-dataset-with-critically-ill-patients
+
+**DESCRIPTION**
+
+To analyze the dataset of Critically Ill Patients and build and train the model on the basis of different features and variables.
+
+There are 2 types of datasets:
+
+- `columns_descriptions`: This dataset contains the description of all the features.
+- `support2`: This dataset contains 9105 entries with 47 different features.
+
+
+### Visualization and EDA of different attributes:
+
+<img alt="heatmap" src="./Images/correlation_heatmap.jpg">
+
+<img alt="graph" src="./Images/age_plot.jpg">
+
+<img alt="graph" src="./Images/diabetes_plot.jpg">
+
+<img alt="graph" src="./Images/dnr_plot.jpg">
+
+<img alt="graph" src="./Images/edu_plot.jpg">
+
+<img alt="graph" src="./Images/temp_plot.jpg">
+
+<img alt="graph" src="./Images/urine_plot.jpg">
+
+
+**MODELS USED**
+
+| Model                     | MSE_train | R2_train | MSE_test  | R2_test   |
+|---------------------------|-----------|----------|-----------|-----------|
+| Random Forest Regression  | 9.52      | 0.88     | 66.02     | 0.14      |
+| XG Boost Regression       | 12.08     | 0.82     | 72.75     | 0.08      |
+| Linear Regression         | 76.54     | 0.07     | 74.91     | 0.06      |
+| Ridge Regression          | 76.54     | 0.07     | 74.91     | 0.06      |
+| Elastic Net Regression    | 77.70     | 0.03     | 75.98     | 0.03      |
+| Decision Tree Regression  | 0.00      | 1.00     | 127.30    | -0.68     |
+| Deep NN                   | 211.83    | -2.24    | 201.13    | -2.35     |
+
+
+**WHAT I HAD DONE**
+
+* Load the dataset which contains 9105 entries in it and having 47 columns in it.
+* Checked for missing values and cleaned the data accordingly.
+* Analyzed the data, found insights and visualized them accordingly.
+* Plotting heatmap using correlation and checking the relation between different features.
+* Found detailed insights of different columns with target variable using plotting libraries.
+* Train the datasets by different models and saves their accuracies into a dataframe.
+
+
+**LIBRARIES NEEDED**
+
+1. Pandas
+2. Matplotlib
+3. Sklearn
+4. NumPy
+5. XGBoost
+6. Tensorflow
+7. Keras
+8. Sci-py
+9. Seaborn
+
+
+**CONCLUSION**
+
+- Random Forest and XG Boost Regression models show promising performance with lower MSE and higher R2 values.
+- Decision Tree Regression achieved perfect R2 on the training set but performed poorly on the test set, indicating overfitting.
+- Deep Neural Network (NN) has a high MSE and negative R2, suggesting poor performance on both training and test sets.
+
+
+**YOUR NAME**
+
+*Avdhesh Varshney*
+
+[![LinkedIn](https://img.shields.io/badge/linkedin-%230077B5.svg?style=for-the-badge&logo=linkedin&logoColor=white)](https://www.linkedin.com/in/avdhesh-varshney-5314a4233/)  [![GitHub](https://img.shields.io/badge/github-%23121011.svg?style=for-the-badge&logo=github&logoColor=white)](https://github.com/Avdhesh-Varshney)
+
diff --git a/Critically Ill Patients Analysis and Prediction/requirements.txt b/Critically Ill Patients Analysis and Prediction/requirements.txt
@@ -0,0 +1,9 @@
+numpy==1.19.2
+pandas==1.4.3
+matplotlib==3.7.1
+scikit-learn~=1.0.2
+scipy==1.5.0
+seaborn==0.10.1
+xgboost~=1.5.2
+tensorflow==2.4.1
+keras==2.4.0