Tools
Tools
Palwe, Yash Sanjay (2021) Hybrid Classifier for Analyzing Customer Satisfaction. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (877kB) | Preview |
Preview |
PDF (Configuration manual)
Download (247kB) | Preview |
Abstract
Sentiment analysis can be used for identifying customer satisfaction from online reviews. Sentiment analysis determines the polarity of the sentence written in natural language whether its positive, negative, and neutral. The challenge is to identify actual opinions about a service or product instead of just determining the polarity of reviews. This research proposes a hybrid classifier for identifying the scale of sentiment at granular level. The hybrid classifier uses techniques such as linguistic feature-based sentence filtering and hybrid feature selection. The model was trained on Amazon Product Reviews dataset. The dataset has 34k reviews. The model was compared based on precision, accuracy and recall and F1-score for analyzing and evaluating customer satisfaction. Results demonstrated that Random Forest with F4 feature outperformed all the models by showing accuracy of 99.5% with 99.54% F1score which is 20% greater than the state of the art. The proposed classifier will be useful for e-commerce organizations for understanding the needs of customer base also product’s shortcomings and best features can be analyzed easily and processing load of business intelligence system can be decreased.
| Item Type: | Thesis (Masters) |
|---|---|
| Subjects: | Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing H Social Sciences > HF Commerce > Electronic Commerce |
| Divisions: | School of Computing > Master of Science in Data Analytics |
| Depositing User: | Tamara Malone |
| Date Deposited: | 27 Feb 2023 16:10 |
| Last Modified: | 01 Mar 2023 17:58 |
| URI: | https://norma.ncirl.ie/id/eprint/6247 |
Actions (login required)
 |
View Item |