Skip to content

Latest commit

 

History

History
27 lines (20 loc) · 795 Bytes

ReadMe.md

File metadata and controls

27 lines (20 loc) · 795 Bytes

Question type classifier

Here we have used the concept of BOW and Term frequency and Inverse document frequency to process the questions and use Logistic regression, Random forest and Naive bayes to train the data and got a F1 score of around 93. Below libraries needs to be installed on computer

  1. Pandas
  2. Numpy
  3. Sklearn
  4. Nltk This is a single file document, and this file don’t require any dependent files expect the above installed file.

methods used

  1. bag of words
  2. term frequency and inverse document frequency
  3. vectorization
  4. Naive Bayes
  5. Logistic Regression
  6. RandomForest

Reference

1. Systems and Approaches for Question Answering: ailao.eu/yodaqa/odbstud.pdf
2. Learning Question Classifiers: www.aclweb.org/anthology/C02-1150