Labels Are a Statistical Estimate, Not a Verdict

It is that no badge or label you see on the screen signifies a definitive truth or a final verdict.

All of the labels like low quality or critical stamped by the app consist of a simple probability calculation. The machine learning model (≈7 MB in size) and the threshold values behind it reflect a mathematically determined probability onto the screen. While the system operates, it can sometimes be wrong and produce incorrect labels from time to time.

Example: The system statistically making a mistake in one out of every ten decisions even in a setting where it operates with ninety percent (0.90) precision.

Go deeper

The decisions of a statistical system depend on the data that system was trained on (36,002 pieces of training data) and the targeted precision floors. Even though the report card values (for example, F1=0.75 in Low Quality TR or F1=0.87 in Critical EN) show the general performance of the system, a margin of error is always present in every individual decision. The app does not tell you the definitive reality of a news item; it merely presents a forecast according to its own internal rules.