{"id":61,"date":"2026-04-03T11:03:53","date_gmt":"2026-04-03T11:03:53","guid":{"rendered":"https:\/\/gigz.pk\/ml\/?post_type=lesson&p=61"},"modified":"2026-04-07T06:42:24","modified_gmt":"2026-04-07T06:42:24","slug":"logistic-regression","status":"publish","type":"lesson","link":"https:\/\/gigz.pk\/ml\/lesson\/logistic-regression\/","title":{"rendered":"\u00a0Logistic Regression"},"content":{"rendered":"\n

Logistic Regression is a widely used Machine Learning algorithm for classification problems<\/strong>. Unlike Linear Regression, which predicts continuous values, Logistic Regression predicts the probability that a given input belongs to a particular class. It is commonly used for binary classification<\/strong> (two classes) but can be extended to multi-class problems.<\/p>\n\n\n\n

How Logistic Regression Works<\/h2>\n\n\n\n

Logistic Regression uses a sigmoid function<\/strong> to map predicted values to probabilities between 0 and 1. The sigmoid function is defined as:<\/p>\n\n\n\n

\u03c3(z) = 1 \/ (1 + e^(-z))<\/code><\/p>\n\n\n\n

Where z<\/code> is the linear combination of input features:<\/p>\n\n\n\n

z = b0 + b1*x1 + b2*x2 + ... + bn*xn<\/code><\/p>\n\n\n\n

The output probability indicates the likelihood that the input belongs to a particular class. A threshold (commonly 0.5) is used to assign the input to a class.<\/p>\n\n\n\n

Steps in Logistic Regression<\/h2>\n\n\n\n
    \n
  1. Collect and Prepare Data:<\/strong> Gather labeled data with input features and binary or categorical output.<\/li>\n\n\n\n
  2. Split Data:<\/strong> Divide the dataset into training and testing sets.<\/li>\n\n\n\n
  3. Train the Model:<\/strong> Fit the logistic regression model to the training data to find the best coefficients.<\/li>\n\n\n\n
  4. Make Predictions:<\/strong> Use the model to predict probabilities and classify new data.<\/li>\n\n\n\n
  5. Evaluate the Model:<\/strong> Use metrics like Accuracy, Precision, Recall, F1 Score, and ROC-AUC to assess performance.<\/li>\n<\/ol>\n\n\n\n

    Applications of Logistic Regression<\/h2>\n\n\n\n