Handwriting Classification of Numbers and Writing Data using the Convolutional Neural Network Model (CNN)

This research delves into the precise classification of handwritten numerical data using a Convolutional Neural Network (CNN) model. A pivotal preprocessing step, namely resizing the dataset images to a consistent dimension, is implemented. This preprocessing enhancement contributes significantly to achieving an impressive classification accuracy of 89.22%. By standardizing image sizes, this crucial preprocessing step ensures uniformity across the dataset, empowering the CNN model to effectively categorize diverse handwritten numbers within the Meteorology, Climatology, and Geophysical Agency (BMKG) Indonesia dataset in 2004. The study underscores the substantial impact of this preprocessing technique in facilitating the CNN model’s robust performance, showcasing its potential in interpreting intricate numerical or weather-related information with a notable accuracy rate of 89.22%.