liveBook · Manning

F-score

As explained in Building Specialized AI Systems

The harmonic mean of precision and recall for a classification system.

F-score is the harmonic mean of precision and recall, providing a balanced measure for classification systems. It is especially useful when urgent cases are rare and both missed cases and false alarms matter. section 1.4.3

Customer-support ticket routing

A fine-tuned GPT-3.5 model classified support tickets using 18 months of historical data. Oversampling minority classes and focal loss helped surface rare security breaches and outages. Critical-incident F1 improved by 60%, while high-priority recall improved by 38%. section 1.4.3

How Building Specialized AI Systems covers it

How deep it goes
Defines the metric and demonstrates outcome improvements, but does not derive its formula or explain implementation internals.
Examples
Uses customer-support tickets, rare security incidents, and high-priority routing as evaluation examples.
What you can do after
Evaluate classification performance by comparing F1 and recall improvements for important ticket classes.
How it connects
Connects classification with precision, recall, fine-tuning, LoRA, oversampling, and focal loss.

Also explained in

Compare how 20 books explain f-score →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Building Specialized AI Systems. The section links point to the parts of the book it draws on.