Machine Learning for Predicting Parliamentary Elections Using Sentiment Analysis | ||||
مجلة الجمعية المصرية لنظم المعلومات وتکنولوجيا الحاسبات | ||||
Volume 38, Issue 38, June 2025, Page 98-108 PDF (4.94 MB) | ||||
Document Type: • البحوث والدراسات والمقالات المستوفاة للقواعد العلمیة المتعارف علیها، والتى یجریها أو یشارک فى إجرائها أعضاء هیئة التدریس والباحثون فى الجامعات ومراکز البحوث المصریة والعربیة، وذلک باللغتین العربیة والإنجلیزیة . | ||||
DOI: 10.21608/jstc.2025.435745 | ||||
![]() | ||||
Authors | ||||
Doaa Alkhiary1; Samir Abu El Fotuoh Saleh2; Mohamed Ebrahim Marie3 | ||||
1Faculty of commerce and Business Administration, Business Information System Department, Helwan University, Egypt | ||||
2Faculty of commerce, Information System, Mansoura University, Egypt | ||||
3Faculty of Computers and Artificial Intelligence, Information System Department, Helwan University, Egypt | ||||
Abstract | ||||
Abstract During the rapid growth of digital technology, the huge increase in digital text data and diverse opinions on social media have created valuable opportunities for innovative sentiment analysis research. This research aims to assess public opinion across various life areas, political polarization, and its role in election campaigns. Twitter, in particular, acts as a rich source of data and public sentiment. This study analyzes a dataset gathered from Twitter’s API, focusing on political opinions related to the 2020 parliamentary elections during both the pre-election period and election day. The dataset includes lists of political parties and independent candidates with Twitter accounts used for campaign promotion. The sample consists of 2,600 tweets, and techniques such as SVM, Naive Bayes (NB), Random Forest (RF), and Decision Trees (DT) were applied along with TF-IDF and weighted averaging to evaluate the results of each method and identify the most accurate one. The comparison revealed that Naive Bayes provided the highest accuracy | ||||
Keywords | ||||
Sentiment Analysis; Machine Learning; Egyptian Parliament Election; algorithms | ||||
Statistics Article View: 67 PDF Download: 25 |
||||