From dd0d919d839f55308c1db861682cbbebe89dd29f Mon Sep 17 00:00:00 2001
From: alolika bhowmik <152315710+alo7lika@users.noreply.github.com>
Date: Mon, 28 Oct 2024 20:52:04 +0530
Subject: [PATCH] Create Program.c

---
 Machine_Learning_Algorithms/SVM/Program.c | 77 +++++++++++++++++++++++
 1 file changed, 77 insertions(+)
 create mode 100644 Machine_Learning_Algorithms/SVM/Program.c

diff --git a/Machine_Learning_Algorithms/SVM/Program.c b/Machine_Learning_Algorithms/SVM/Program.c
new file mode 100644
index 00000000..493afd72
--- /dev/null
+++ b/Machine_Learning_Algorithms/SVM/Program.c
@@ -0,0 +1,77 @@
+#include <stdio.h>
+#include <stdlib.h>
+
+#define LEARNING_RATE 0.01
+#define EPOCHS 1000
+#define LAMBDA 0.01 // Regularization parameter
+
+// Sample data structure for a data point
+typedef struct {
+    double x1;
+    double x2;
+    int label;
+} DataPoint;
+
+// Function to initialize weights
+void initializeWeights(double *weights, int size) {
+    for (int i = 0; i < size; i++) {
+        weights[i] = 0.0;
+    }
+}
+
+// Function to calculate dot product
+double dotProduct(double *weights, DataPoint point) {
+    return weights[0] * point.x1 + weights[1] * point.x2;
+}
+
+// SVM training function using Stochastic Gradient Descent
+void trainSVM(DataPoint *data, int dataSize, double *weights) {
+    for (int epoch = 0; epoch < EPOCHS; epoch++) {
+        for (int i = 0; i < dataSize; i++) {
+            DataPoint point = data[i];
+            double y = point.label;
+            double prediction = dotProduct(weights, point);
+
+            // Update rule for SVM
+            if (y * prediction < 1) {
+                weights[0] += LEARNING_RATE * ((y * point.x1) - (2 * LAMBDA * weights[0]));
+                weights[1] += LEARNING_RATE * ((y * point.x2) - (2 * LAMBDA * weights[1]));
+            } else {
+                weights[0] += LEARNING_RATE * (-2 * LAMBDA * weights[0]);
+                weights[1] += LEARNING_RATE * (-2 * LAMBDA * weights[1]);
+            }
+        }
+    }
+}
+
+// Function to make predictions
+int predict(double *weights, DataPoint point) {
+    double prediction = dotProduct(weights, point);
+    return (prediction >= 0) ? 1 : -1;
+}
+
+int main() {
+    // Training data (XOR dataset)
+    DataPoint data[] = {
+        {2, 3, 1},
+        {1, 1, -1},
+        {2, 1, -1},
+        {3, 2, 1},
+        {3, 3, 1},
+        {1, 2, -1}
+    };
+    int dataSize = sizeof(data) / sizeof(data[0]);
+    
+    double weights[2];
+    initializeWeights(weights, 2);
+
+    // Train the SVM model
+    trainSVM(data, dataSize, weights);
+
+    // Test the SVM model
+    DataPoint testPoint = {3, 3, 1};
+    int prediction = predict(weights, testPoint);
+    printf("Prediction for test point (3, 3): %d\n", prediction);
+
+    return 0;
+}