Text Classification Model

This repository contains a Python project for binary text classification using a feedforward neural network. The model is implemented using the Keras library and processes text data to classify it into two categories: positive and negative.

Features

• Data preprocessing:
  • Text cleaning and tokenization.
  • Removal of punctuation and stopwords.
  • Conversion of text data to frequency-based matrices.
• Neural network:
  • Input layer with 50 neurons and ReLU activation.
  • Output layer with sigmoid activation for binary classification.
  • Optimized with the Adam optimizer.
• Visualized model architecture as an image (model.png).

Requirements

• Python 3.x
• Libraries:
  • NumPy
  • Keras
  • TensorFlow
  • NLTK

How to Run

1. Clone the repository.

   git clone https://github.com/KPlanisphere/text-classification-model.git
   cd text-classification-model

2. Install dependencies.

   pip install -r requirements.txt

3. Place the dataset in the appropriate folders (NEGATIVO and POSITIVO).
4. Run the script.

   python lab10.py

File Structure

• lab10.py: Main script for training and evaluating the model.
• vocabulario.py: Script to manage vocabulary processing.
• vocabulario.txt: Contains the vocabulary used in the dataset.

Results

After training, the script will output the model's accuracy on the test set. The trained model's structure is saved as model.png.