Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/476161Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Nazlia Omar, Prof. Dr. | |
| dc.contributor.author | Adel Qasem Abdo Al-Shabi (P56111) | |
| dc.date.accessioned | 2023-10-06T09:14:12Z | - |
| dc.date.available | 2023-10-06T09:14:12Z | - |
| dc.date.issued | 2013-07-11 | |
| dc.identifier.other | ukmvital:74680 | |
| dc.identifier.uri | https://ptsldigital.ukm.my/jspui/handle/123456789/476161 | - |
| dc.description | Sentiment analysis (SA) deals with the computational treatment of opinion, sentiment, and subjectivity in text. It is a rapidly growing research area due to the explosive growth of the user-generated content. SA is a challenging interdisciplinary task that contains natural language processing, web mining and machine learning. Most of the studies on Sentiment Analysis only deal with English documents, perhaps due to the lack of resources in other languages. Despite the fact that Arabic is currently among the top ten languages most used on the Internet according to the Internet World Stats, there are very few resources for managing sentiments or opinions such as sentilexicons and opinion corpora. The main obstacle is that phrases and words that are used by Arabic web users to express sentiments are highly subjected to usage trends. In addition, the use of dialectal phrases and words contributes to ambiguity in the analysis of sentiments and opinions. In order to remedy this deficiency, this research proposes a two-level ensemble of machine learning classifiers framework for handling the problem of subjectivity and sentiment analysis for Arabic customer reviews. In the first level, a subjectivity analyser based on ensemble of machine learning classifiers is used to filter the reviews. The ensemble of the classifiers includes Naive Bayes, Ngram and Rocchio classifier. In the second level, a sentiment analyser classifies relevant reviews into the three categories i.e. positive, negative and neutral. The experimental results show that the ensemble of the classifiers improves the classification effectiveness in terms of macro-F1 for both levels. The best results obtained for the subjectivity analysis and the sentiment classification in terms of macro-F1 are 0.895 and 0.905 respectively. The experimental results show that our proposed method is able to classify Arabic customers' reviews more accurately than the traditional methods.,Master/Sarjana | |
| dc.language.iso | eng | |
| dc.publisher | UKM, Bangi | |
| dc.relation | Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat | |
| dc.rights | UKM | |
| dc.subject | Subjectivity and sentiment analysis | |
| dc.subject | Arabic customers | |
| dc.subject | Ensemble classification-based approach | |
| dc.subject | Natural language processing (Computer science) | |
| dc.title | A two-level ensemble classification-based approach for subjectivity and sentiment analysis of Arabic customers reviews | |
| dc.type | theses | |
| dc.format.pages | 84 | |
| dc.identifier.callno | QA76.9.N38.S483 2013 3 | |
| dc.identifier.barcode | 000832 | |
| Appears in Collections: | Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| ukmvital_74680+Source01+Source010.PDF Restricted Access | 1.39 MB | Adobe PDF |  View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.