Sentiment analysis with Andrew Trask. The main focus of this project is to understand how to create a neural network that can anlyse a review and classify it as bad or good. It is split into several parts:
- curating a data set.
- training a a neural network
- increasinghe signal and reducing the noise in teh data set.
- Download anaconda
conda create --name sentiment_alysis python=3
- Activate the source
source activate sentiment_alysis
- Ensure you have numpy, jupyter notebook, matplotlib, scikit-learn, and bokeh installed by doing the following:
conda install numpy matplotlib jupyter notebook scikit-learn bokeh
- Run the following to open up the notebook:
jupyter notebook Sentiment\ Classification.ipynb
You can open the jupyter notebook directly in github Sentiment Classification.ipynb
