Handwritten Digit Recognition using Machine Learning

For my semester project, I developed a handwritten digit recognition system using machine learning. The project involved training a machine learning algorithm to recognize handwritten digits and then using that model to predict the digits in test images. I used the MNIST dataset for training and testing the model, which consists of 60,000 training images and 10,000 testing images.

To develop the system, I first preprocessed the data by normalizing the images and converting them to grayscale. Then, I used the K-nearest neighbors (KNN) algorithm to train the model on the training data. After fine-tuning the model, I achieved an accuracy of 96.5% on the testing data, which demonstrates the effectiveness of the approach.

Overall, the project allowed me to gain experience with machine learning algorithms, data preprocessing techniques, and the Python programming language. It was an exciting opportunity to apply theoretical knowledge to a practical problem and develop a system that can potentially be used for various applications, such as digit recognition in banking and postal services.