Skip to content

Supervised Linear Model(SVM) Evaluation with Sparse(TFIDF) and dense(Word2Vec) text representations

Notifications You must be signed in to change notification settings

AimVoma/Accord.NET-SVM

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 

Repository files navigation

Accord.NET-SVM Repo

Project Description

Machine Learning Application that performs Sentiment Classification(Fine-Grained, emotions) with sparse text representations(TFIDF) or pre-trained dense word vectors(Word2Vec), on Supervised Linear Model(SVM). The Classification result is later dumped in local storage as a Confusion Matrix(CM) Analysis. The implementation of the project was based on Accord.NET, a Machine Learning Framework written completely in Csharp for production-grade application development.

App.Config
  • Basic Configurations of IO operations
  • Text Representations(TFIDF-W2V)
  • SVM model setup
Forms of Text Representations
  • TFIDF - Sparse
  • Word2Vec(W2V) - Dense
Forms of Linear Optimization Function/Solver
  • SMO(Sequential Minimization Optimization)
  • LCD(Linear Coordinate Descent)
Prerequisite:Accord.NET

About

Supervised Linear Model(SVM) Evaluation with Sparse(TFIDF) and dense(Word2Vec) text representations

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages