Predicting Human Behavior Using Arabic Sentiment Analysis on Social Media: Approaches and Challenges | ||
| النشرة المعلوماتية في الحاسبات والمعلومات | ||
| Volume 6, Issue 1 - Serial Number 20240601, January 2024, Pages 22-33 PDF (997.32 K) | ||
| Document Type: المقالة الأصلية | ||
| DOI: 10.21608/fcihib.2023.201708.1084 | ||
| Authors | ||
| محمد عبده* 1; سيد عبد الجابر1; مروة صلاح فرحان1, 2; نرمين عثمان1, 2 | ||
| 1قسم نظم المعلومات ، كلية الحاسبات والذكاء الاصطناعي ، جامعة حلوان ، القاهرة ، مصر | ||
| 2كلية المعلوماتية وعلوم الحاسب، الجامعة البريطانية في مصر، القاهرة، مصر | ||
| Abstract | ||
| The rapid growth of social networking and micro-blogging websites has motivated researchers to analyze published content to identify and predict human behavior. With the ongoing growth in data volume, the efficient and effective extraction of valuable information has become crucial. Researchers are tackling this problem using big data analytics. However, most studies have focused on the English language and fewer research efforts have been devoted to the Arabic language. This paper covers the recent Arabic sentiment analysis research, highlighting the most important studies that analyzed content from social media to predict human behavior and the different approaches used. Sentiment analysis studies varied between lexicon-based, traditional machine learning-based, deep learning, and hybrid approaches, in addition to employing swarm intelligence to optimize the performance of text classification algorithms. The reviews showed that Naïve Bayes (NB) and Support Vector Machine (SVM) were the most widely used algorithms among traditional machine learning algorithms. The results also showed that using deep learning approaches achieves better accuracy than other approaches. Also, the use of optimization algorithms based on swarm intelligence had a significant impact on increasing the accuracy of text clustering. | ||
Highlights | ||
| ||
| Keywords | ||
| Sentiment Analysis; Human Behavior; Machine Learning; Natural Language Processing; Deep Learning | ||
| Full Text | ||
|
| ||
| References | ||
|
| ||
|
Statistics Article View: 314 PDF Download: 224 |
||
