“Classifying social media bots as malicious or benign using semi-supervised machine learning” from the Journal of Cybersecurity

This article highlights the importance of understanding the behavior of social media bots and the impact they have on online communication and social dynamics. The theory being tested is whether machine learning techniques can accurately classify social media bots as either malicious or benign using a semi-supervised approach. A hypothesis that can be made is that the machine learning model can be trained to accurately identify patterns and characteristics of social media bots and distinguish between those that are malicious and those that are benign, even when only partially labeled data is available. This relates to sociology because it talks about the nature of the study of social media and how important it is to gain insight from the field of computer science, data analysis, and social sciences to help understand the evolving online communication. This topic raises questions about the role of technology that shapes the values of society. Because social media provides great insight for their interaction with users impacting social media. There is a lot of research talked about in this article; specifically data analyzing and collection. Semi-supervised machine learning in short terms consists of some labeled data and a large amount of unlabeled data. This data is used to make multiple predictions. The authors collected data from the social media bots from a real-world social network. The use of this machine learning and data analysis provided a great understanding of the behavior of social media bots, contributing to the classification of them. However, of course there are several challenges and concerns that marginalized groups encounter. The main one is correctly and accurately classifying the social media robots. Inaccurate classification can lead to large consequences for these groups that will be targeted by the bots. They can be unfairly targeted as well. However, this article provides insight for others and society to contribute and develop policies and technology to protect these groups. Specifically, learning and creating algorithms will significantly help. This can relate to the topics talked about in class because of how much the social media bot can affect the behavior of a group or individual. In class, we have learned a lot about the behavior of an individual in technological terms. The authors created a great article on social media bots with the use of semi-supervised machine learning, providing information that can lead to implications and more findings for future research. Overall, the study used a combination of network-based and content-based features to train a machine learning model that effectively classified social media bots with high accuracy.The results of this study indicate that semi-supervised machine learning is a promising technique for detecting social media bots and could potentially improve the security and safety of social media platforms.

Reference: Innocent Mbona, Jan H P Eloff, Classifying social media bots as malicious or benign using semi-supervised machine learning, Journal of Cybersecurity, Volume 9, Issue 1, 2023, tyac015, https://doi.org/10.1093/cybsec/tyac015