Sentiment Analysis — Methodology

We track how Kenyans react to Parliament, transparently. This page documents exactly how the analysis works and which sources we use.

How It Works

• Collection. We ingest articles from vetted mainstream outlets and comments on their official YouTube channels, every few hours.
• Topic tagging. Each item is matched to policy topics (e.g. Healthcare, Education, Economy) so sentiment can be broken down by issue.
• Classification. Each item is classified as positive, neutral or negative with a score from −1.0 to +1.0 by a large language model, with conservative handling of sarcasm and mixed sentiment.
• Aggregation. Scores are averaged by topic, by source and over time. Everything on the dashboard is filterable by date.

Which Sources We Use

Analysis is drawn exclusively from reputable mainstream media — established national newspapers and the official channels of established TV and radio newsrooms. This curated whitelist keeps the analysis anchored to professional, accountable journalism. The current whitelist:

• Citizen Digital (News website)
• Citizen TV Kenya (YouTube channel)
• KTN News Kenya (YouTube channel)
• NTV Kenya (YouTube channel)
• Nation (News website)
• The Standard (News website)
• The Star (News website)

Topics Tracked

• Agriculture
• Corruption & Accountability
• Devolution
• Economy & Taxation
• Education
• Governance & Constitution
• Healthcare
• Security

Sentiment is indicative, not definitive. We welcome corrections via Contact.