⬟ Sentiment & AI
Machine-Readable Market Data: From Candlesticks to Market State
Machine-readable market data turns 200 years of candlestick knowledge into executable intelligence — how Oyamori reads market states instead of charts.
The Oyamori Thesis™: From Predicting Markets to Measuring Market Energy
Market energy trading measures whether price will move — not which way. Oyamori reads momentum, volume, and volatility expansion to time options entries.
"News Sentiment Trading: How to Profit Before the Market Reacts"
Learn how news sentiment trading works, how AI scores headlines before prices move, and how to build a signal-to-trade pipeline that acts faster than the crowd.
Black-Box AI vs. Transparent Sentiment
Black-box AI signals are unverifiable. A signal you cannot inspect is one you cannot validate — and a signal you cannot validate is a risk you cannot manage.
Building a News-Aware Algo with Newsvibe API
End-to-end tutorial: authenticate with Newsvibe API, parse sentiment signals, define entry logic, execute trades via Alpaca, and log results — all in Python.
How Newsvibe Works
Newsvibe is a sentiment engine you can inspect and validate. Here is how raw news becomes a scored, structured trading signal with tier classification.
News-Driven Gap Trading
Gap trading meets sentiment scoring — overnight news predicts the morning gap direction if you weight recency, urgency, and volume correctly.
Why News Sentiment Changes Everything in Algo Trading
News sentiment is not a soft signal — it is a structural information advantage. Here is why sentiment-blind algos fail around high-impact news events.
Regime Detection with Sentiment
Sentiment regime detection is the best signal for when not to trade — aggregate negative sentiment across a broad basket marks risk-off conditions reliably.
Sentiment + Momentum: Signal Fusion in Practice
Signal fusion cuts false positives: one signal has noise, two calibrated uncorrelated signals have less. Here is the implementation and the math.
Signal Fusion Explained
Signal fusion cuts false positives: a single signal has noise, two calibrated uncorrelated signals have less. Here is the math and the architecture behind it.