Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/476452Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Saidah Saad, Dr. | |
| dc.contributor.author | Bilal Sabri Diekan Al-Gebory (P75876) | |
| dc.date.accessioned | 2023-10-06T09:18:44Z | - |
| dc.date.available | 2023-10-06T09:18:44Z | - |
| dc.date.issued | 2016-09-07 | |
| dc.identifier.other | ukmvital:96933 | |
| dc.identifier.uri | https://ptsldigital.ukm.my/jspui/handle/123456789/476452 | - |
| dc.description | Sentiment Analysis (SA) is the process of detecting the contextual polarity of text. It determines whether a piece of writing is positive, negative or neutral. Most of the current studies related to this field of research focus mainly on English texts with very limited resources available for texts in other languages like the Arabic language. Several approaches have been proposed to tackle the Arabic sentiment analysis. However, these algorithms require a large amount of computing power. Additionally, this research deals with a few Arabic dialects. All these difficulties add to the limited of languages resources serve to motivate more research work in this area. As a result, the main aim of this research study is to propose a technique for sentiment analysis in Arabic social media. This technique used a Modern standard language (MSA) and Dialect Arabic (DA). As well, this research study emphasizes on document-level sentiment analysis using the Naive Bayes classifier as a Meta classifier that combines two methods namely Association Rule (AR) mining and the N-gram model. In addition, three feature selection methods (FSM) were used with this technique are Information Gain (IG), Chi-square (CHI) and Gini Index (GI). For the purpose of this research, the corpus used is Opinion Corpus of Arabic (OCA) corpus that consists of 500 Arabic reviews. Finally, the result shows that the combined classifier used in this research study reported an accuracy of F-Measure (90.8%) which has outperformed the baseline. This means that proposed model is suitable for the implementation of sentiment analysis of reviews and comments from the social media networks.,Certification of Master's/Doctoral Thesis" is not available | |
| 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 | Sentiment analysis | |
| dc.subject | Arabic language | |
| dc.subject | Social media | |
| dc.subject | Universiti Kebangsaan Malaysia -- Dissertations | |
| dc.title | Sentiment feature analysis in Arabic social media using association rule and N-gram model | |
| dc.type | theses | |
| dc.format.pages | 105 | |
| dc.identifier.barcode | 002829(2017) | |
| Appears in Collections: | Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| ukmvital_96933+SOURCE1+SOURCE1.0.PDF Restricted Access | 242.58 kB | Adobe PDF |  View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.