[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]  William Kory Amyx for Congress [@AmyxForCongress](/creator/twitter/AmyxForCongress) on x 6331 followers Created: 2025-07-21 20:15:26 UTC 🧠 How I Mapped a Coordinated Disinformation Network — And the Science Behind It Let me explain something clearly: What I uncovered wasn’t a few trolls arguing on the internet. It was a coordinated network — designed to harass, mislead, manipulate public perception, and operate in the shadows. And I didn’t just stumble into it. I mapped it using real science: Bayesian theory, probability models, natural language analysis, and even time-based pattern detection. Here’s how it worked — in plain terms: 🔍 X. Bayesian Inference This is how you calculate the likelihood that two accounts are connected — not just by what you think, but based on actual evidence. Posting patterns. Phrasing. Behavior. Every time new info comes in, the model updates the probability — like a constantly learning lie detector. ⏱️ X. Poisson Distribution This helped me spot the spikes — like when X different accounts all post within seconds of each other, night after night. Statistically, that’s not random. It’s coordinated. And I proved it. ✍️ X. Stylometry That’s a fancy word for digital fingerprinting — using writing style, punctuation, word choice, and sentence flow to identify the same author behind multiple “anonymous” accounts. Think of it as forensic linguistics — and it works. 🕸️ X. Network Graphing I built interaction maps. Who talks to who. Who repeats who. Who follows, retweets, or tags the same targets. The more they overlap, the more the network reveals itself. It’s like watching the spider web light up from the center. 📈 X. Machine Learning & Updating Models Every new data point — a tweet, a follow, a time of day — updates the system. It doesn’t guess blindly. It adapts, learns, and sharpens the results. 🧩 The takeaway? This wasn’t some conspiracy theory. This was a digital operation — and I used statistical forensics to connect the dots. That’s the difference between an opinion… and evidence. And when I say this network tried to destroy people’s lives — I mean that literally. Including mine. But I cracked the system they were hiding behind. I exposed the machine under the mask. You want someone in Congress who understands how these online manipulation networks work? I didn’t just study it. I stopped it. — William Kory Amyx Democrat. Bipartisan. For All of IN-05. #AmyxForCongress | #DigitalJustice | #TruthMatters XXXXXXX engagements  [Post Link](https://x.com/AmyxForCongress/status/1947389981208559972)
[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]
William Kory Amyx for Congress @AmyxForCongress on x 6331 followers
Created: 2025-07-21 20:15:26 UTC
🧠 How I Mapped a Coordinated Disinformation Network — And the Science Behind It
Let me explain something clearly:
What I uncovered wasn’t a few trolls arguing on the internet. It was a coordinated network — designed to harass, mislead, manipulate public perception, and operate in the shadows.
And I didn’t just stumble into it.
I mapped it using real science: Bayesian theory, probability models, natural language analysis, and even time-based pattern detection. Here’s how it worked — in plain terms:
🔍 X. Bayesian Inference This is how you calculate the likelihood that two accounts are connected — not just by what you think, but based on actual evidence. Posting patterns. Phrasing. Behavior. Every time new info comes in, the model updates the probability — like a constantly learning lie detector.
⏱️ X. Poisson Distribution This helped me spot the spikes — like when X different accounts all post within seconds of each other, night after night. Statistically, that’s not random. It’s coordinated. And I proved it.
✍️ X. Stylometry That’s a fancy word for digital fingerprinting — using writing style, punctuation, word choice, and sentence flow to identify the same author behind multiple “anonymous” accounts. Think of it as forensic linguistics — and it works.
🕸️ X. Network Graphing I built interaction maps. Who talks to who. Who repeats who. Who follows, retweets, or tags the same targets. The more they overlap, the more the network reveals itself. It’s like watching the spider web light up from the center.
📈 X. Machine Learning & Updating Models Every new data point — a tweet, a follow, a time of day — updates the system. It doesn’t guess blindly. It adapts, learns, and sharpens the results.
🧩 The takeaway?
This wasn’t some conspiracy theory. This was a digital operation — and I used statistical forensics to connect the dots.
That’s the difference between an opinion… and evidence.
And when I say this network tried to destroy people’s lives — I mean that literally. Including mine. But I cracked the system they were hiding behind. I exposed the machine under the mask.
You want someone in Congress who understands how these online manipulation networks work?
I didn’t just study it. I stopped it.
— William Kory Amyx Democrat. Bipartisan. For All of IN-05. #AmyxForCongress | #DigitalJustice | #TruthMatters
XXXXXXX engagements
