# Accuracy, macro-F1 and confusion matrices

Source: https://onco.cc/terms/accuracy-f1/  
OnCo record `accuracy-f1` (Term). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

Accuracy is the share of predictions that were right; macro-F1 averages the F1 score of each class equally, so a model cannot look good by getting only the common cancer types right.

## Summary

The F-score is a measure of predictive performance for binary classification calculated from precision and recall (Wikipedia). For a multi-class task such as cancer-type prediction, macro-F1 averages per-class F1 with equal weight, which exposes failures on rare classes that overall accuracy hides; a confusion matrix shows which classes are mixed up, often tissues of shared origin. Accuracy on a task saturated by an easy signal (tissue identity) leaves no room to show a new modality helps, a ceiling effect.

## Fields

- Kind: Term
- Last checked: 2026-09-24
- Also known as: accuracy metric; macro-F1; macro F1; F1 score; F-score; F-measure; confusion matrix; balanced accuracy; classification accuracy
- Tags: cansim-terms

## Notes

- Listed in the CanSim terms map 1.0.0 (docs/onco/terms.json, generated 2026-09-24), CC BY 4.0, attribution: CanSim project, an open, public-data-first cancer foundation-model programme; CanSim page path /terms/accuracy-macro-f1.

## Sources

- Wikipedia: https://en.wikipedia.org/wiki/F-score
- Wikipedia: accuracy and precision: https://en.wikipedia.org/wiki/Accuracy_and_precision
- Wikipedia: https://en.wikipedia.org/wiki/F-score

## Connected records

- terms: [Ablation study and multi-task heads](https://onco.cc/terms/ablation-study/), [Cancer AI vocabulary (CanSim terms map)](https://onco.cc/terms/cancer-ai-vocabulary/), [ROC-AUC, PR-AUC and time-dependent AUC](https://onco.cc/terms/roc-auc/), [Tissue-of-origin signal in tumour data](https://onco.cc/terms/organ-of-origin-signal/)

---
JSON: https://onco.cc/api/v1/entities/accuracy-f1.json