Fisher's linear discriminant

E212221

dimensionality reduction method feature extraction method linear classifier statistical technique supervised learning method

Fisher's linear discriminant is a classic statistical technique for dimensionality reduction and classification that projects data onto a line to maximize separation between classes.

Try in SPARQL Jump to: Surface forms Disambiguation Statements Elicitation Referenced by

All labels observed (3)

Label	Occurrences
Fisher's discriminant	1
Fisher's linear discriminant canonical	1
Fisher's linear discriminant analysis	1

How this entity was disambiguated

This entity first appeared as the object of triple T1908300 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.

NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07

Target entity: Fisher's linear discriminant
Context triple: [Ronald A. Fisher, knownFor, Fisher's linear discriminant]

A. PCA
The PCA is an intergovernmental organization based in The Hague that facilitates arbitration and other forms of dispute resolution between states, state entities, intergovernmental organizations, and private parties.
B. PCA
PCA (Principal Component Analysis) in scikit-learn is a dimensionality reduction technique that transforms high-dimensional data into a smaller set of uncorrelated components capturing the most variance.
C. PCA
PCA is a conservative evangelical Presbyterian denomination in the United States known for its adherence to Reformed theology and Presbyterian church governance.
D. Hotelling’s T-squared distribution
Hotelling’s T-squared distribution is a multivariate generalization of Student’s t-distribution used primarily for hypothesis testing and constructing confidence regions for mean vectors in multivariate statistics.
E. LogisticRegression
LogisticRegression is a scikit-learn machine learning estimator that models the probability of class membership using a linear decision boundary with logistic (sigmoid) or related link functions.
F. None of above. chosen
G. Unsure - the case is ambiguous/there is not enough information to decide.

NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07

Target entity: Fisher's linear discriminant
Target entity description: Fisher's linear discriminant is a classic statistical technique for dimensionality reduction and classification that projects data onto a line to maximize separation between classes.

A. PCA
PCA (Principal Component Analysis) in scikit-learn is a dimensionality reduction technique that transforms high-dimensional data into a smaller set of uncorrelated components capturing the most variance.
B. PCA
The PCA is an intergovernmental organization based in The Hague that facilitates arbitration and other forms of dispute resolution between states, state entities, intergovernmental organizations, and private parties.
C. PCA
PCA is a conservative evangelical Presbyterian denomination in the United States known for its adherence to Reformed theology and Presbyterian church governance.
D. Hotelling’s T-squared distribution
Hotelling’s T-squared distribution is a multivariate generalization of Student’s t-distribution used primarily for hypothesis testing and constructing confidence regions for mean vectors in multivariate statistics.
E. LogisticRegression
LogisticRegression is a scikit-learn machine learning estimator that models the probability of class membership using a linear decision boundary with logistic (sigmoid) or related link functions.
F. None of above. chosen

Statements (50)

Predicate	Object
instanceOf	dimensionality reduction method ⓘ feature extraction method ⓘ linear classifier ⓘ statistical technique ⓘ supervised learning method ⓘ
alsoKnownAs	Fisher's discriminant ⓘ linked to: Fisher's linear discriminant Fisher's linear discriminant analysis ⓘ linked to: Fisher's linear discriminant
assumes	Gaussian class-conditional distributions for optimality ⓘ class labels are known during training ⓘ classes are linearly separable in projected space ⓘ equal covariance matrices across classes for classical derivation ⓘ
basedOn	between-class covariance ⓘ class means ⓘ within-class covariance ⓘ
category	classification algorithm ⓘ
comparedTo	logistic regression ⓘ principal component analysis ⓘ support vector machines ⓘ
criterion	ratio of between-class scatter to within-class scatter ⓘ
differsFrom	principal component analysis by using class labels ⓘ
generalization	multi-class linear discriminant analysis ⓘ
goal	maximize between-class variance ⓘ maximize separation between classes ⓘ minimize within-class variance ⓘ
limitation	performance degrades with highly overlapping classes ⓘ sensitive to small sample size relative to dimensionality ⓘ
mapsFrom	high-dimensional feature space ⓘ
mapsTo	one-dimensional subspace ⓘ
mathematicalForm	generalized eigenvalue problem ⓘ
namedAfter	Ronald A. Fisher ⓘ
operatesOn	labeled data ⓘ
optimizationObjective	Rayleigh quotient of scatter matrices ⓘ
output	discriminant direction vector ⓘ scalar discriminant scores for samples ⓘ
projectionType	linear projection ⓘ
property	invariant to non-singular linear scaling of features ⓘ
relatedTo	linear discriminant analysis ⓘ
requires	estimation of scatter matrices ⓘ
typicalUseCase	two-class classification ⓘ
usedFor	classification ⓘ dimensionality reduction ⓘ pattern recognition ⓘ supervised feature projection ⓘ
usedIn	biometrics ⓘ face recognition ⓘ machine learning ⓘ medical diagnosis ⓘ pattern recognition ⓘ signal processing ⓘ statistics ⓘ

How these facts were elicited

Referenced by (3)

Full triples — surface form annotated when it differs from this entity's canonical label.

Ronald A. Fisher → knownFor → Fisher's linear discriminant ⓘ

Fisher's linear discriminant → alsoKnownAs → Fisher's discriminant ⓘ

linked to: Fisher's linear discriminant

Fisher's linear discriminant → alsoKnownAs → Fisher's linear discriminant analysis ⓘ