object

knn_classifier

k-Nearest Neighbors classifier with multiple distance metrics and weighting options. Learns from a dataset object implementing the dataset_protocol protocol and returns a classifier term that can be used for prediction and exported as predicate clauses.

Availability:
logtalk_load(knn_classifier(loader))
Author: Paulo Moura
Version: 2:0:0
Date: 2026-05-11
Compilation flags:
static, context_switching_calls
Imports:
public probabilistic_classifier_common
Uses:
format
list
pairs
type
Remarks:
(none)
Inherited public predicates:
 check_classifier/1  check_option/1  check_options/1  classifier_options/2  default_option/1  default_options/1  diagnostic/2  diagnostics/2  export_to_clauses/4  export_to_file/4  learn/2  learn/3  option/2  option/3  predict/3  predict_probabilities/3  print_classifier/1  valid_classifier/1  valid_option/1  valid_options/1

Public predicates

predict/4

Predicts the class label for a new instance using the learned classifier and the given options. The instance is a list of Attribute-Value pairs.

Compilation flags:
static
Template:
predict(Classifier,Instance,Class,Options)
Mode and number of proofs:
predict(+compound,+list,-atom,+list(compound)) - one

predict_probabilities/4

Predicts class probabilities for a new instance using the learned classifier and the given options. Returns a list of Class-Probability pairs. The instance is a list of Attribute-Value pairs.

Compilation flags:
static
Template:
predict_probabilities(Classifier,Instance,Probabilities,Options)
Mode and number of proofs:
predict_probabilities(+compound,+list,-list,+list(compound)) - one

Protected predicates

(no local declarations; see entity ancestors if any)

Private predicates

(no local declarations; see entity ancestors if any)

Operators

(none)

See also

dataset_protocol, c45_classifier, isolation_forest_anomaly_detector, naive_bayes_classifier, nearest_centroid_classifier, random_forest_classifier, adaptive_boosting_classifier