Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/463371Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Nazlia Omar, Prof. Dr. | |
| dc.contributor.author | Ahmed Abdulkareem Shayah (P65625) | |
| dc.date.accessioned | 2023-09-25T09:23:27Z | - |
| dc.date.available | 2023-09-25T09:23:27Z | - |
| dc.date.issued | 2014-12 | |
| dc.identifier.other | ukmvital:77084 | |
| dc.identifier.uri | https://ptsldigital.ukm.my/jspui/handle/123456789/463371 | - |
| dc.description | Recently the Web became the primary sources for conveying the sentiment and exchange views about products and other goods and services. The opinions and experiences of other people constitute an important source of information in our everyday life. The sentiment analysis in Arabic language is a difficult task due to the lack of Arabic language resources, and the variety of dialects in Arabic. Although many studies used supervised and unsupervised learning in Arabic document-level but rarely studies have focused on the feature-level sentiment in spite of its importance in Arabic language. In this research, a hybrid method is proposed which combines lexicon-based approach and a machine learning approach to extract the product features and to determine the opinions' polarities of these features. It takes advantage of the sentiment dictionary, i.e. Arabic senti-lexicon to analyze the opinion orientation expressed on the product features. To evaluate our method, a data set which consists of positive and negative opinion reviews are used. This annotated dataset contains reviews for three products. It is also annotated with product features that appeared in the reviews. In the experiments, we particularly compared three classifiers i.e. K-Nearest Neighbor (KNN), Naive Bayes (NB) and Support Vector Machine (SVM) with sentiment lexicon. Overall, the results of experiments show that SVM classifier that has been combined with sentiment lexicon gave the best results with an accuracy of 94%. In general, the results of the experiment suggest that the hybrid learning method can be an effective method in mining product features for sentiment analysis.,Master of Information Technology | |
| dc.language.iso | eng | |
| dc.publisher | UKM, Bangi | |
| dc.relation | Faculty of Science and Technology / Fakulti Sains dan Teknologi | |
| dc.rights | UKM | |
| dc.subject | Lexicology | |
| dc.subject | Computational linguistics | |
| dc.title | A hybrid lexicon-based and machine learning approach for Arabic product review | |
| dc.type | theses | |
| dc.format.pages | 70 | |
| dc.identifier.callno | QA76.5913.S973 2014 3 tesis | |
| dc.identifier.barcode | 000954 | |
| Appears in Collections: | Faculty of Science and Technology / Fakulti Sains dan Teknologi | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| ukmvital_77084+SOURCE1+SOURCE1.0.PDF Restricted Access | 1.53 MB | Adobe PDF |  View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.