Skip to content

A simple movie review semantic classifier using scikit-learn

License

Notifications You must be signed in to change notification settings

abrigham1/MovieReviewClassifier

Repository files navigation

Movie Review Classifier

The purpose of this package is to perform semantic analysis on movie reviews using machine learning. It makes use of the imdb review dataset and is based on some excellent tutorials from Sebastian Raschka in Python Machine Learning.

Note model was originally persisted using scikit-learn version 0.18.1 so if your version is different you may need to re-persist the model by running

path/to/python path/to/MovieReviewClassifier/MovieReviewClassifier.py

To run this you will need to install python and many scientific packages the easiest way to get all that is by installing Anaconda

To run semantic predictions just run

path/to/python path/to/MovieReviewClassifier/MovieReviewPredict.py "Movie Review text goes here"

Output follows the format Prediction,Probability for example positive,93.5 if the algorithm is 93.5% sure it is a positive review

With settings as is it is around 90% accurate feel free to further tune it on your own.

About

A simple movie review semantic classifier using scikit-learn

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages