Natural Language Processing, or NLP for short, has been an active research field for more than 50 years. It is defined as the automatic manipulation of natural language, speech or text, by software.
In this project, I created a Neural Network Model to identify insincere questions - those founded upon false premises, or that intend to make a statement rather than look for helpful answers - on the website Quora, that provided the dataset in a Kaggle competition.
The project is divided in two notebooks: Quora kernel shows the preprocessing and training phase. Analysis one shows posterior analysis on the predicted results and the whole data.
The training data was not included in this repository due to its size. However, it can still be found on the competition page:
https://www.kaggle.com/c/quora-insincere-questions-classification
- Python 3.7
- Numpy
- Scikit-learn
- TensorFlow
- Pytorch