Text Classification using naive bayes

Predicting Amazon ratings based on reviews by Text Classification using the Naive Bayes Algorithm.

Name

naive - Run the executable program for Naive Bayes

Synopsis

./naive <part> <tr> <ts> <output>

Description

This program will train naive bayes model using given code on train data, make predictions on test data and write final predictions in given output file.

Options

• PART

  Part i.e. a,b or c

• TR

  File containing training data in csv format where 1st entry is the target

• TS

  File containing test data in csv format where 1st entry is the target

• OUT

  Output file (write your predictions in this file)

Example

./naive a train.csv test.csv output

Parts

• Part A

  • Used Unigrams as features
  • Used Laplace Smoothing to avoid zero probabilities(c = 1)
  • Used Logarithms to avoid underflow issues
  • Implemented everything from the First Principles and not usde any existing R/python modules.

• Part B

  • All in Part A plus some Pre-Processing
  • Stopword Removal(nltk)
  • Stemming(nltk)
  • String Cleaning

• Part B

  • Feature Engineering
  • Lemmatization
  • Used Bigrams as features

Data

• amazon_train.csv: Train data
• amazon_test_public.csv: Public Test data

Note: In the Public test data, actual class labels are replaced with -1