All Classes and Interfaces
Class
Description
Common attributes and methods for dataset readers.
Abstract class for defining experiments.
Abstract class defining common variables and methods for likelihood-based scores.
Abstract class defining common variables and methods for any kind of node.
Contains common attributes and methods for PGM.
Contains common attributes and methods for classes that determine the structure constraints of PFG.
This class provides a method to compute the average local log-likelihood of each node of a Multi-CTBNC.
Implements a Bayesian network (BN).
Implements the Bayesian estimation to estimate the parameters of a BN.
Implements the Bayesian Dirichlet equivalence metric for Bayesian networks with nodes that have CPTs (Heckerman et
al., 1995).
Implements hill climbing algorithm for BNs.
Implements the maximisation phase (hill climbing algorithm) of the hybrid structure learning algorithm for Bayesian
networks.
Implements the hybrid structure learning algorithm for Bayesian networks.
Stores the parameter and structure learning algorithms for a Bayesian network.
Implements the log-likelihood score for Bayesian networks with nodes that have CPTs.
Maximum likelihood estimation of BN parameters.
Defines methods for parameter learning algorithms of discrete Bayesian networks.
Builds the specified parameter learning algorithm for a BN.
Interface used to define scores for Bayesian networks.
Compute and store the sufficient statistics of a discrete BN node.
Implements the tabu search algorithm for Bayesian networks.
Extends the DiscreteNode class to store a CIM and the sufficient statistics for a CTBN.
Service that creates and manages a
ClassificaionTask.Task that allows executing the classification of sequences in a background thread.
Interface representing classification models.
Provides static methods for the creation of classifiers.
Abstract class representing a concept drift adaptive method.
This class implements a concept drift adaptive method that operates globally on a MultiCTBNC model.
This class implements a concept drift adaptive method that operates locally on each node of a MultiCTBNC model.
Interface representing scores that can be computed locally on each node of a MultiCTBNC and used to detect concept
drifts in a given data batch.
This class implements the ConInd online feature subset selection algorithm from Yu et al. 2018.
Allows writing the results of the experiments through the standard output stream.
Controller used to initialise the elements of the GUI and allow the interaction between the logic of the application
and the GUI.
Utility class with methods related to controlling the UI behaviour.
Extends the DiscreteNode class to store a CPT and the sufficient statistics for a BN.
Implements a cross-validation method used to learn one CTBNC for each class variable and merge the results.
Implements cross-validation method.
Implements a continuous-time Bayesian network (CTBN).
Bayesian parameter estimation for a discrete CTBN.
Implements the Bayesian Dirichlet equivalence metric for CTBNs with nodes that have CIMs (Nodelman et al., 2003).
Implements the conditional log-likelihood score for Multi-CTBNCs with nodes that have CPTs and CIMs to define its
bridge and feature subgraph (represented by a CTBN).
Implements hill climbing algorithm for CTBNs.
Implements the maximisation phase (hill climbing algorithm) of the hybrid structure learning algorithm for
continuous-time Bayesian networks.
Implements hill climbing algorithm for CTBNs.
Implements the hybrid structure learning algorithm for continuous-time Bayesian networks.
Stores the parameter and structure learning algorithms for a continuous-time Bayesian network.
Implements the log-likelihood score for CTBNs with nodes that have CIMs.
Maximum likelihood estimation of CTBN parameters.
Define methods for parameter learning algorithms of continuous-time Bayesian networks.
Builds the specified parameter learning algorithm for a CTBN.
Interface used to define scores for continuous-time Bayesian networks.
Computes and stores the sufficient statistics of a discrete CTBN node.
Implements the tabu search algorithm for continuous-time Bayesian networks.
Implementation of the CTPC algorithm for Multi-CTBNCs.
Implements the restriction phase (CTPC algorithm) of the hybrid structure learning algorithm.
Defines the restrictions of a general directed acyclic graph.
Implements a Multi-CTBNC where the class subgraph is formed by a Bayesian network, while the feature subgraph is a
K-dependence continuous-time Bayesian network, i.e., the feature nodes are limited to have K parents (apart of the
class variables).
Implements methods for the generation and writing of datasets sampled from Multi-CTBNCs.
Represents a time series dataset, which stores sequences and provides methods to access and modify their
information.
Interface for classes that read datasets.
Creates dataset readers.
Represents an experiment for evaluating continuous-time Bayesian network classifiers on streaming data.
The class is designed for reading and processing streaming data from multiple CSV files.
Specifies the structure restrictions of a CTBN.
Abstract class defining common variables and methods for discrete nodes.
Implements a Multi-CTBNC with an empty class subgraph.
Implements a Multi-CTBNC with an empty class subgraph and a K-dependence continuous-time Bayesian network for the
feature subgraph, i.e., the feature nodes are limited to having K parents (apart of the class variables).
It only allows the creation of empty BNs.
Thrown when a valid sequence could not be created with the provided data.
Thrown when an error occurs due to an incorrect value provided by the user.
Service that creates and manages an
EvaluationTask.Task that allows executing the training and evaluation of a model in a background thread.
Allows writing the results of the experiments in an Excel file.
Represents an experiment that can be executed.
A factory class for creating instances of experiments based on provided arguments.
Implements an experiment where a feature stream is treated as a static dataset.
Represents an experiment for evaluating continuous-time Bayesian network classifiers on feature streams.
Factory class for creating specific types of feature stream experiments.
Abstract class representing an implementation of an experiment with feature streams.
Class responsible for reading multiple CSV files representing a feature stream.
Represents an experiment that processes a feature stream with online feature subset selection using a MultiCTBNC.
Represents an experiment for processing a feature stream dataset with online feature subset selection without
updating the model.
Represents an experiment for processing a feature stream dataset without online feature subset selection but with
model updates.
Implements first-choice Hill Climbing.
Implements common attributes and methods for hill climbing algorithms.
Defines an interface for different implementations of the hill climbing algorithm.
Class used to contain the solution given by the hill climbing algorithms.
Implementation of the HITON-PC algorithm.
Implements hold-out validation method.
This class implements a experiment on streaming data.
JavaFX application to interact with the CTBNLab software.
This class serves as the entry point for the data stream sampling application.
This class represents the JavaFX application for data stream sampling.
Main class for running experiments.
This class serves as the entry point for the feature stream sampling application.
This class represents the JavaFX application for feature stream sampling.
Class to sample datasets from Multi-CTBNCs with provided or randomly generated structures.
Implementation of the MB-CTPC algorithm.
Implements the structure constraints of a Max-k continuous-time Bayesian network classifier, i.e., a CTBNC where the
number of parents of the feature nodes is bounded by a positive number (Codecasa and Stella, 2014).
Interface used to be able to pass evaluation metrics as parameters of other methods.
Computes different metrics for the evaluation of multi-dimensional classifications.
Defines classes that write the results of evaluation metrics on different outputs.
Represents an experiment for comparing different models' performance on datasets with different settings.
Implements the multi-dimensional continuous-time Bayesian network classifier (Multi-CTBNC).
Specifies the structure constraints of a multidimensional continuous-time naive Bayes classifier (Multi-CTBNC) where
any subgraph has arcs except the bridge subgraph (fully naive multi-dimensional classifier).
Provides the logic to read separate CSV files.
Manages the writing of datasets into CSV files
Defines the structure of a continuous-time Naive Bayes classifier.
Thrown when a state was never seen before by the classifier.
Interface for a generic node of a PGM.
Provides static methods for the creation of nodes.
Links nodes with a unique index.
Thrown when a requested task is not implemented.
This interface defines the structure for classes that implement online feature subset selection algorithms.
This class extends the MB-CTPC algorithm to an online learning context, allowing for dynamic updates to the
learned model based on new data.
Implements the Page Hinkley Test for concept drift detection.
Interface for parameter learning algorithms.
Implementation of the PC algorithm discrete-state Bayesian networks.
Implements the restriction phase (PC algorithm) of the hybrid structure learning algorithm.
Defines the methods of a probabilistic graphical model (PGM)
Contains a multidimensional prediction and its probability.
Utility class with methods related to the estimation of probabilities.
Implements random-restart hill climbing.
Represents a sequence of multivariate data, i.e., a set of data points with multiple variables where the order is
relevant.
Reads time series data contained in a single CSV.
Represents a sliding window data structure.
Represents the state of certain nodes/variables (events) by keeping their names and values.
Provides the basis for statistical-based feature subset selection algorithms.
Interface used to define classes that specify structure constraints for PGMs.
Interface used to define algorithms for learning the structure of PGMs.
Builds the specified structure learning algorithms for Bayesian networks and continuous-time Bayesian networks.
Class designed to perform a comparative experiment of structure learning algorithms for Multi-CTBNCs.
Class that encapsulates the names of the feature variables selected by a FSS algorithm and the execution time of this
algorithm to provide that solution.
Interface for sufficient statistics of discrete nodes.
Implements a validation method for evaluating CTBNCs using a test.
This class allows specifying different training and test datasets.
A class for creating and managing a time series chart using the JFreeChart library.
Service that creates and manages a
TrainingTask.Task that allows executing the training of a model in a background thread.
Class to write the predictions made on a dataset in a TXT file.
Thrown when the provided dataset could not be read.
Utility class with methods related to the user interface.
Utility class.
Abstract class defining common methods for validation algorithms.
Builds validation methods.
Thrown when an expected variable is not found in a provided dataset.
A class for creating and managing an XY line chart using the JFreeChart library.