Proposed Machine-Learning Aided Framework For Misinformation Management In Uganda Using Word 2Vector And Long Short-Term Memory: A Case Study Of Uganda Communications Commision
Loading...
Date
2023-11
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Kampala International University
Abstract
Information quality is becoming an increasingly important issue as social networks have become the primary source for misinformation dissemination. Because of their ease of use, spreading behaviour, and low cost, social network platforms are leveraging news consumption. Many studies have been developed on methods to improve fake news classification, particularly on misinformation detection on social media, with promising results in recent years. In Uganda, information filtering was still done in a traditional way hence not efficient and effective. To address this challenge, this study aims at providing a framework that will manage misinformation using the machine learning algorithms like word2vec and LSTM. The study found that the most common type of misinformation is fake news and hate speech that is spread by Ordinally users and politician as the main agents that spread misinformation via social media platforms like TikTok and twitter respectively. I also extracted data from Twitter using Apache NIFI, elastic search and Kibana for data visualization. The dataset information included, tweets, the user details, retweets, tweet URL and date. Subsequently, I performed an offline analysis through the use of machine learning and deep learning techniques. I hope that this research work will provide useful insights for realizing ever more effective tools to counter misinformation and those who spread it intentionally.
Description
A report submited in partial fulfilment for the award of degree of master of science in information systems to
school of mathematics and computing (somac) Kampala International University