Customer Reviews Sentiment Analysis: A hybrid technique of Lexicon and Machine Learning based Classification model (SVM, NB, Logistic Regression)

Bhalerao, Komal Vijay (2021) Customer Reviews Sentiment Analysis: A hybrid technique of Lexicon and Machine Learning based Classification model (SVM, NB, Logistic Regression). Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration manual) Download (2MB) | Preview

Abstract

The majority of items are available online in our digital age. E-commerce platforms are evolving in order to put products within the reach of online users in order to deliver the highest level of customer pleasure and convenience of use. People nowadays tend to rely on feedback before ordering any goods online, therefore reading hundreds of evaluations takes up a lot of time for customers. Making decisions on enhancing quality of the product and acquiring insights, companies and organizations can obtain lot of data from customer sentiment analysis. A lot of research has previously been implemented on the classification of Sentiment Analysis based on many different aspects and techniques, however, not a lot of research has been done with a combination of Lexicon based and Machine Learning classification model. The process of Sentiment analysis can be tedious since the data available is textual format and it is the most unstructured type of data available. In this research, to enable efficient and outstanding outcome for classification, text pre-processing is carried out and two types of feature extractors are used. In order to fulfil this task, three machine learning models were implemented. The outcome generated by these models were evaluated using different evaluation matrices and the results were compared. SVM provided the best accuracy for classification i.e. 91% using TF-IDF vector.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	LP; Sentiment Analysis; Machine Learning; TF-IDF; CountVectorizer; SVM; Naive Bayes; Logistic Regression
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software T Technology > T Technology (General) > Information Technology > Computer software H Social Sciences > HF Commerce > Electronic Commerce
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Clara Chan
Date Deposited:	15 Nov 2021 10:29
Last Modified:	15 Nov 2021 10:29
URI:	https://norma.ncirl.ie/id/eprint/5135

Actions (login required)

View Item