shammy-lp/kenya-tweets

# Kenya Tweet Classifier: Detecting Real-time Hate Speech and Misinformation Data science - Fulltime Remote Timeline: 19th March - 6th May # Project By: Foresight Analytica ## Group Members 1.Morgan Abukuse 2.Dennis Mwania 3.Linet Patriciah 4.Precious Kalia 5.Felista Kiptoo # Introduction ## Project Overview In this project, we take a look at the increasing challenge of hate speech and misinformation on Kenyan social media, particularly on platforms like Twitter. With the rapid rise in internet and smartphone access, millions of Kenyans now engage online daily. While this has opened up new channels for communication, it has also created room for the spread of harmful content, often targeting individuals or groups based on tribe, gender, political affiliation, or religion. This growing issue poses real social threats, especially during elections, public unrest, or national discussions where misinformation and hate can escalate tensions, incite violence, or spread fear. To help tackle this, we have developed a real-time tweet classification system that uses natural language processing (NLP) and machine learning techniques to automatically detect and flag tweets containing hate speech or misinformation. We aim to build a system that helps moderate content in real-time, supports fact-checking efforts, and promotes a safer digital space for all users in Kenya. By identifying toxic content early, this project supports peace-building, public awareness, and more responsible use of social media. ## Bussiness Problem In Kenya, Twitter has become a powerful space where people share opinions and talk about national issues. It gives citizens a voice and helps them organize around important topics like politics, health, and education. However, this freedom has also led to problems. There’s been a rise in hate speech and misinformation, especially during elections or times of political tension. Tweets that attack certain tribes, spread false news, or incite violence can …