Design, Improvement, Development, and Performance Analysis of a Collection of Model Developed From Naïve Bayes and Maximum Entropy Opinion Mining Classifiers for Movie Reviews
About this article
Abstract
The internet is a basic platform for people from every one walks of life to interconnect and convey opinions on the topic of their choice. Almost every website asks for comments, suggestions and reviews. Exploring opinion and determining a person’s views is itself a large subject in computer science, known as Opinion Mining, also called Sentiment Analysis There are different sentiment classifiers, the most admired of which are the Naïve Bayes classifier, maintain Vector Machines (SVM), Maximum Entropy classifier, to name a few. In this paper, here we are analyzing the efficient performance of the Naïve Bayes also about the Maximum Entropy classifiers. Here we analyze and examine how bigrams perform better than unigrams in sentiment analysis. We further propose a serialized ensemble model of the two as a hybrid algorithm and analyze its performance as well.
Keywords
- Reviews
- measure
- Sentiment analysis
- Naïve Bayes
- Maximum Entropy
- bigrams
- ensemble model
- hybrid algorithm
References
[1] Wikipedia.org ‘Intenet Movie Database’, 2015 [Onlne]. Available:https://en.wikipedia.org/wiki/Internet_Movie_Database
[2] Jayashri Khairnar, Mayura Kinikar, ‘Machine Learning Algorithms for Opinion Mining and Sentiment Classification’, International Journal of Scientific and Research Publications, Volume 3, Issue 6, June 2013
[3] Nltk.org ‘NLTK 3.0 Documentation’, [Online]. Available: http://www.nltk.org/
[4] Wikipedia.org ‘Precision and recall’, 2015 [Online]. Available: https://en.wikipedia.org/wiki/Precision_and_recall
[5] Wikipedia.org ‘Bag-of-words model’, 2015 [Online]. Available: https://en.wikipedia.org/wiki/Bag-of-words_model
View more references (16)
[6] Wikipedia.org ‘Naive Bayes Classifier’, 2015, [Online]. Available: https://en.wikipedia.org/wiki/Naive_Bayes_classifier
[7] Nlp.stanford.edu ‘Naïve Bayes text classification’, 2008, [Online] Available: http://nlp.stanford.edu/IR-book/html/htmledition/naive-bayes-text-classification-1.html
[8] Wikipedia.org ‘Entropy (information theory)’, 2015 [Online]. Available: https://en.wikipedia.org/wiki/Entropy_(information_theory)
[9] Wikipedia.org ‘Principle of maximum entropy’, 2015 [Online]. Available: https://en.wikipedia.org/wiki/Principle_of_maximum_entropy
[10] Bayes.wustl.edu ‘Information Theory and Statistical Mechanics’ 1957, [Online] Available: http://bayes.wustl.edu/etj/articles/theory.1.pdf
[11] Sentiment.christopherpotts.net, ‘Sentiment Symposium Tutorial: Classifiers’ 2011, [Online], Available: http://sentiment.christopherpotts.net/classifiers.html#maxent
[12] Wikipedia.org ‘Bigram’, 2015, [Online] Available: https://en.wikipedia.org/wiki/Bigram
[13] Streamhacker.com ‘TEXT CLASSIFICATION FOR SENTIMENT ANALYSIS – ELIMINATE LOW INFORMATION FEATURES, 2010, [Online]. Available: http://streamhacker.com/tag/feature-extraction/
[14] Wikipedia.org ‘Ensemble learning’, 2015, [Online], Available: https://en.wikipedia.org/wiki/Ensemble_learning
[15] AnalyticsVidhya.com, ‘Basics of Ensemble Learning Explained in
[16] Simple English’, 2015,[Online],Available: http://www.analyticsvidhya.com/blog/2015/08/introduction-ensemble-learning/
[17] ‘The Optimality of Naive Bayes’, Harry Zhang, [Online], Available: http://www.aaai.org/Papers/FLAIRS/2004/Flairs04-097.pdf
[18] Wikipedia.org ‘Confusion Matrix’, 2015, [Online], Available: https://en.wikipedia.org/wiki/Confusion_matrix
[19] Data School ‘Simple guide to confusion matrix terminology’, 2014,[Online], Available http://www.dataschool.io/simple-guide-to-confusion-matrix-terminology/
[20] Christine Day, ‘The Importance of Sentiment Analysis in Social Media
Analysis’, [Online], Available: