The Observation Theory EncyclopediaFrom TSKAboutBy kindBy chapterBy Lean fileLedgerProvenance

naive Bayes instrument

DefinitionA classifier that multiplies one-dimensional likelihoods and adds their logs, so that each feature's contribution is the same whatever the others are and it cannot read an interaction. Chapter 6.
ExampleTwo features with likelihood ratios 3 and 2 give a combined ratio of 6, whatever the other features are.
BookData Mining as Observation, draft 0.2, commit f3914f0; entry id naive-bayes, kind instrument.
Statusno ledger row names this entry. Corrections: none recorded.
Defining equationnone
Assumptions and scope
  • A classifier that multiplies one-dimensional likelihoods and adds their logs. The log of the product is the sum of the logs, the log odds add across features, and one feature’s contribution is the same whatever the others are, so the score has no interaction terms and its Hessian is diagonal.
  • What naive Bayes cannot read is the interaction, which is a limit of the model class rather than a direction in its kernel, and its read operator is still not diagonal in general because two sensitivities can rise and fall together across rows.
Prior artnone recorded
Evidencelean/DataMiningAsObservation/NaiveBayes.lean
Reviewednot yet reviewed; generated 2026-09-10 from records at the commits on the provenance page.
no interaction terms
One-dimensional likelihoods multiplied, so the score is additive and the Hessian diagonal.

Equation

none

Conditions

Conditions are curated in entries.toml rather than read from a record.

Ledger

none

First stated

Chapter 6 section 6.1 of Data Mining as Observation, after TSK chapter 4.

Measurements

none

Failures and corrections

none

Invariance envelope

none declared

Machine checked

lean/DataMiningAsObservation/NaiveBayes.lean, theorems log_likelihood_sum, log_odds_sum, contribution_independent, at observation-data-mining f3914f0; what the check covers is stated in the book’s appendix C.

Used in

Data Mining as Observation chapters 6.

Related

classifier; Hessian; confidence; logistic regression.

See also

Book equations stated beside the entry’s terms, not defining it: 6.1, 0.9.

Sources-table rows that share a record with the entry without naming it: chapter 2 section 2.3, chapter 6 section 6.1.

Status

Generated 2026-09-10 by encyclopedia/generate.py; book at observation-data-mining f3914f0; the commit of every record is listed in the encyclopedia’s provenance.

← Nadeau and Bengio correctionnat →