https://github.com/pierremtb/imdb-binary-sentiment-analysis

Course project at McGill University. Implementation from scratch of a Bernouilli Naïve Bayes classifier and text-processing features.
https://github.com/pierremtb/imdb-binary-sentiment-analysis

data-science machine-learning mcgill-university python3 sentiment-analysis text-classification

Last synced: 10 months ago
JSON representation

Course project at McGill University. Implementation from scratch of a Bernouilli Naïve Bayes classifier and text-processing features.

• Host: GitHub
• URL: https://github.com/pierremtb/imdb-binary-sentiment-analysis
• Owner: pierremtb
• License: mit
• Created: 2019-02-23T10:32:47.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2019-02-23T11:11:57.000Z (over 7 years ago)
• Last Synced: 2025-02-11T23:56:52.886Z (over 1 year ago)
• Topics: data-science, machine-learning, mcgill-university, python3, sentiment-analysis, text-classification
• Language: Python
• Size: 33.6 MB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# IMBD Binary Sentiment Analysis

Course project (COMP551) at McGill University, to classify movie reviews straight out of [IMDb](https://www.imdb.com/) as positive or negative. 

Implementation from scratch in Python 3 of a [Bernouilli Naïve Bayes](https://en.wikipedia.org/wiki/Naive_Bayes_classifier#Bernoulli_naive_Bayes) classifier and text-processing features, mainly lexicons and n-grams.

## Performance

The course hosted a [Kaggle](https://www.kaggle.com/) competition.

This model as-is, without external Machine Learning libraries got the following score on the test set.

![F1 Score](https://latex.codecogs.com/gif.latex?F_1=&space;0.86693)

## Lexicon reference

>Minqing Hu and Bing Liu. "[Mining and Summarizing Customer Reviews](https://www.cs.uic.edu/~liub/publications/kdd04-revSummary.pdf)."

Proceedings of the ACM SIGKDD International Conference on Knowledge 

Discovery and Data Mining (KDD-2004), Aug 22-25, 2004, Seattle, 

Washington, USA

## Copyright

[MIT license](LICENSE.md)