Social media has permeated our lives such that a majority of unstructured communications happen through these platforms. Social media provides a ready audience anytime. As a result, a large portion of discourses that were initially articulated verbally have shifted to social media platforms. Control of communication over social media is difficult to achieve or enforce considering that operators of the social media platforms are foreign companies operating from different legislative environments compared to those of their users. This problem is exacerbated by the extreme fast flow and dynamism of information on social media. Therefore, sensitive and potentially harmful content can propagate through social media platforms quickly without being detected. Hate speech is an example of such communication that can be difficult to deal with particularly when it is propagated through social media platforms. Actually, Mr. António Guterres – the United Nations (UN) Secretary-General – describes Social media as a global megaphone for hate.
Propagation of hate-speech over social media platform is relatively a new phenomenon considering that social media platforms are recent disruptive technologies. Hate speech can be propagated on social media through: text messages, pictures, videos, emoji, or emoticons. Sometimes the hate speech could be obfuscated in online content that seem ordinary. Various approaches including intentional misspelling by swapping characters, elongating words using many repeated letters, or putting spaces between letters have been used to obfuscate hate speech in social media discourses [17]. Hiding of hate text messages in images and hate sarcasm using images or videos are other techniques of hiding hate speech over social media platforms.
It is challenging to deal with hate speech that is propagated over social media platforms as compared to that which is propagated over traditional media like newspapers, magazines, TV, radio or billboards. Unlike in traditional media, online hate speech can be produced and distributed easily, at low cost and anonymously while having the potential to reach a global and diverse audience in real time. The relative permanence of online content is also problematic when hateful discourse can resurface and regain popularity over time. The UN concludes that efforts of understanding and monitoring the dynamics of hate speech across diverse online communities and platforms often stall given the sheer scale and diversity of the phenomenon, current technological limitations of automated monitoring systems and the opacity of online companies. If hate-speech propagation over social media platforms is not tamed, it has potential of completely destroying the social, political and economic stability of nations.
The main objective of this project is to develop “Darubini”, an application for detecting political hate speech that is propagated over social media platforms in Kenya. Darubini is a Swahili word for telescope i.e., an optical instrument for observing distant astronomical objects like stars. Searching for hate-speech over social media platforms is synonymous to searching for new heavenly objects among the many stars in sky.
Investigators
- Dr Malanga Ndenga, Kirinyaga University
- Dr Mageto Mokaya, Kirinyaga University
- Kenneth Chesoli, Dedan Kimathi University of Technology