Multilayer Perceptron & Feedforward Neural Networks for Text and Data Classification

Multilayer Perceptron & Feedforward Neural Networks for Text and Data Classification

Developed and implemented multiple neural network models using Multilayer Perceptron (MLP) and Feedforward Neural Networks (FFNN) for classification tasks.

The project explored neural network architectures, training using backpropagation and gradient descent, and applying models to structured and text datasets.

Key work included building models in Python and R, performing data preprocessing and feature scaling, and training networks with different hidden layer configurations.

The models were applied to datasets such as Iris classification, SMS spam detection, and 20 Newsgroups text classification, demonstrating how neural networks can learn complex patterns for predictive tasks.

Technologies: Python, R, Scikit-learn, TensorFlow, Keras, Weka, Neural Networks, Machine Learning.