Skip to main content


Browse the glossary using this index

Special | A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z | ALL

Page: (Previous)   1  2  3  4  5  6  7  8  9  10  ...  15  (Next)


Black box model

A type of machine learning model whose representation is complex, not directly understandable or readable by the user, and where the inference procedure used to determine the output is unknown.


Type of ensemble that uses weights or costs for examples that are more difficult to identify correctly. In this case, M iterations are performed, with each one generating an estimator dependent on the result of the previous stage.



Instance representing a set or cluster of instances.

Chord diagram or Dependency wheel

Allows the visualization of weighted relationships between several entities. The graph has a circular shape, where each entity is represented by a fragment on the outside of the circular arrangement. The size of each of these fragments reflects the frequency at which that entity occurs. The fragments of different entities are connected by bands or links where the width or size of the band corresponds to the frequency of occurrence of that combination of entities in the data set.


Name given to the task of making class or category label predictions for new samples.

Classification paradigm

Each of the different classification model types according to the type of discriminant function considered.

Cloud computing

Cloud computing allows us to work with computing and storage servers on a network, usually over the internet.


A set of instances that resemble each other.

Cluster analysis

An approach used to analyze the RAs obtained. It consists of analyzing the rules by grouping them according to the items they contain. We can analyze them by creating groups of rules with some common element in the antecedent or consequent. Groups can also be created by selecting rules that have the same itemset in the consequent and antecedent, allowing the study of different rules with related associations.


These are unsupervised learning techniques whose objective is to group available objects (instances) so that objects within the same group (cluster) are like each other and are different from objects in other groups.

Page: (Previous)   1  2  3  4  5  6  7  8  9  10  ...  15  (Next)