Diabetes Predictor

We will use the following Python libraries:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

# Load the data
diabetes_data = pd.read_csv(r'C:\Users\Francesco\Desktop\Data Project\DIABETE\diabetes.csv')

diabetes_data.shape

diabetes_data.head()

# Check for missing values in the dataset
missing_values = diabetes_data.isnull().sum()
print("\nMissing values for each column:")
print(missing_values)

# Generate summary statistics for the dataset
summary_stats = diabetes_data.describe()
print("\nSummary statistics for the dataset:")
print(summary_stats)

# Plot histograms for each column in the dataset BEFORE data cleaning
plt.figure(figsize=(20, 12))
for i, column in enumerate(diabetes_data.columns, 1):
    plt.subplot(3, 3, i)
    sns.histplot(diabetes_data[column], kde=True, bins=30)
plt.tight_layout()
plt.title("Data distributions BEFORE cleaning")
plt.show()

# Identify columns with unrealistic zero values
columns_with_zeros = ['Glucose', 'BloodPressure', 'SkinThickness', 'Insulin', 'BMI']

for column in columns_with_zeros:
    count_zeros = (diabetes_data[column] == 0).sum()
    print(f"Number of zero entries in {column}: {count_zeros}")

# Replace zero values with NaN for the identified columns
for column in columns_with_zeros:
    diabetes_data[column].replace(0, np.nan, inplace=True)

# Impute NaN values with the median of the respective columns
for column in columns_with_zeros:
    diabetes_data[column].fillna(diabetes_data[column].median(), inplace=True)

print("\nData after cleaning and preprocessing:")
print(diabetes_data.head())

# Plot histograms for each column in the dataset AFTER Data Cleaning
plt.figure(figsize=(20, 12))
for i, column in enumerate(diabetes_data.columns, 1):
    plt.subplot(3, 3, i)
    sns.histplot(diabetes_data[column], kde=True, bins=30)
plt.tight_layout()
plt.show()

# Boxplots for Each Feature (Excluding 'Outcome')
# ------------------------------------
plt.figure(figsize=(20, 15))

# Exclude 'Outcome' from the columns list
features_for_boxplot = diabetes_data.columns[:-1]

# Plotting boxplots for the selected features
for i, column in enumerate(features_for_boxplot, 1):
    plt.subplot(3, 3, i)
    sns.boxplot(y=diabetes_data[column])
    plt.title(f"Boxplot of {column}")

plt.tight_layout()
plt.show()

# Correlation Heatmap

correlation_matrix = diabetes_data.corr()

plt.figure(figsize=(12, 9))
sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm', linewidths=0.5, fmt=".2f")
plt.title("Correlation Heatmap")
plt.show()

Glucose & Outcome: There’s a moderate correlation of 0.47, suggesting that glucose levels have a noticeable relationship with the diabetes outcome.
Age & Pregnancies: There’s a moderate correlation of 0.54, indicating that as age increases, the number of pregnancies tends to increase as well.
BMI & SkinThickness: There’s a moderate correlation of 0.39, suggesting a relationship between body mass index and skin fold thickness.
Most other correlations are relatively low, suggesting limited linear relationships between them. However, remember that correlation doesn’t imply causation, and non-linear relationships won’t be captured well by this metric.

# Distribution of Glucose Levels based on Diabetes Outcome
plt.figure(figsize=(10, 6))
sns.kdeplot(data=diabetes_data, x='Glucose', hue='Outcome', fill=True, common_norm=False, palette="coolwarm")
plt.title('Distribution of Glucose Levels by Diabetes Outcome')
plt.xlabel('Glucose Levels')
plt.ylabel('Density')
plt.legend(title='Diabetes Outcome', labels=['Diabetes', 'No Diabetes'])
plt.show()

# Import necessary libraries for model building
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier

# Separate features and target variable from the dataset
X = diabetes_data.drop('Outcome', axis=1)
y = diabetes_data['Outcome']

# Split the data into training and testing sets
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Initialize the Random Forest classifier
rf_model = RandomForestClassifier(n_estimators=100, random_state=42)

# Train the classifier using the training data
rf_model.fit(X_train, y_train)

I Chose the Random Forest classifier, a robust learning method known for its high accuracy and ability to handle large datasets.
Split the data into training and test sets (80-20), and trained the model.

# Define sample data for a young woman and an older man
young_woman = [[1, 90, 70, 25, 80, 22, 0.3, 25]]
older_man = [[0, 140, 85, 30, 130, 30, 0.5, 60]]

# Use the trained model to make predictions for the samples
young_woman_prediction = rf_model.predict(young_woman)
older_man_prediction = rf_model.predict(older_man)

# Display the predictions
print("\nSample Predictions:")
print("Prediction for young woman:", "Diabetic" if young_woman_prediction[0] == 1 else "Non-diabetic")
print("Prediction for older woman:", "Diabetic" if older_man_prediction[0] == 1 else "Non-diabetic")

# Import necessary libraries for model evaluation
from sklearn.metrics import accuracy_score, precision_score, recall_score, roc_auc_score

# Make predictions on the test set using the trained classifier
rf_predictions = rf_model.predict(X_test)

# Display performance metrics for the predictions
print("\nRandom Forest Model Performance Metrics:")
print("Accuracy:", accuracy_score(y_test, rf_predictions))
print("Precision:", precision_score(y_test, rf_predictions))
print("Recall:", recall_score(y_test, rf_predictions))
print("AUC-ROC:", roc_auc_score(y_test, rf_predictions))

# ------------------------------------
# Visualize Confusion Matrix for Random Forest Model
# ------------------------------------
from sklearn.metrics import confusion_matrix

# Generate the confusion matrix
conf_matrix = confusion_matrix(y_test, rf_predictions)

# Visualize the confusion matrix using a heatmap
plt.figure(figsize=(8, 6))
sns.heatmap(conf_matrix, annot=True, cmap='Blues', fmt='g')
plt.xlabel('Predicted')
plt.ylabel('Actual')
plt.title('Confusion Matrix for Random Forest Model')
plt.show()

The Random Forest model’s performance on the test data is summarized as follows:

---