Skip to content

Applies the Expectation Maximization algorithm on semi-supervised Rotten Tomatoes data, classifying sentences as either positive or negative reviews

Notifications You must be signed in to change notification settings

MarkPreschern/Semi-Supervised-Rotten-Tomatoes

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

19 Commits
 
 
 
 
 
 

Repository files navigation

Semi-Supervised-Rotten-Tomatoes

Applies the Expectation Maximization algorithm on semi-supervised Rotten Tomatoes data, classifying sentences as either positive or negative reviews

To alter performance, provide the following program arguments (optional):

  • --semiSupervised <True/False> Whether to consider the semi-supervised data in the file, or to do a completely unsupervised run
  • --lemmatize <True/False> Whether to lemmatize the input sentences
  • --fixedSeed <True/False> Whether to perform the algorithm on a fixed seed for Random
  • --iterations <Positive Integer> The number of iterations the EM algorithm will perform
  • --topWords <Positive Integer> The number of words/bigrams that are printed per class
  • --naiveBayes <True/False> Whether to use the Naive Bayes model or the Markov model of bigrams

To run the program from the console, type py RottenTomatoesClassifier <optional program arguments> and ensure that trainEMsemisup.txt is in the same directory as RottenTomatoesClassifier.py

About

Applies the Expectation Maximization algorithm on semi-supervised Rotten Tomatoes data, classifying sentences as either positive or negative reviews

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages