A Probabilistic Graphical Model-Based Approach for the Label Ranking Problem

Alfaro, Juan C.; Rodrigo, Enrique González; Aledo, Juan A.; Gámez, José A.

doi:10.1007/978-3-030-29765-7_29

Cited by 5 publications

(7 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, we describe the scheme proposed in the preliminary (conference) version of this study [ 23 ], based on the learning process used in [ 36 ] and which basically wraps the EM method for parameter estimation. In Algorithm 1, we show our adaptation from the NB estimation algorithm [ 31 ] to the LR problem.…”

Section: Hidden Naive Bayes For Label Rankingmentioning

confidence: 99%

“…The results obtained in [ 23 ] shed light on certain drawbacks. The main one is that the algorithm reaches the stopping condition too soon, which results in a small number of components for the mixture.…”

Section: Hidden Naive Bayes For Label Rankingmentioning

confidence: 99%

“…To overcome the constraints regarding the topology of the network when dealing with different types of variables, in the preliminary version of this study, we proposed a mixture-based structure where the root is a hidden discrete variable . In [ 23 ], we based our proposal on a Naive Bayes graphical structure, where only univariate probability distributions are estimated for each state of the hidden variable. Learning and inference schemes were also designed in [ 23 ], based on the use of well-known Expectation-Maximization (EM) algorithm for parameter estimation and a combination of probabilistic inference with the Kemeny Ranking Problem ( KRP ) [ 24 ], respectively.…”

Section: Introductionmentioning

confidence: 99%

“…In [ 23 ], we based our proposal on a Naive Bayes graphical structure, where only univariate probability distributions are estimated for each state of the hidden variable. Learning and inference schemes were also designed in [ 23 ], based on the use of well-known Expectation-Maximization (EM) algorithm for parameter estimation and a combination of probabilistic inference with the Kemeny Ranking Problem ( KRP ) [ 24 ], respectively. Nonetheless, the proposed methods performed somewhat unevenly when dealing with the different datasets.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Mixture-Based Probabilistic Graphical Models for the Label Ranking Problem

Rodrigo

Alfaro

Aledo

et al. 2021

Entropy

View full text Add to dashboard Cite

The goal of the Label Ranking (LR) problem is to learn preference models that predict the preferred ranking of class labels for a given unlabeled instance. Different well-known machine learning algorithms have been adapted to deal with the LR problem. In particular, fine-tuned instance-based algorithms (e.g., k-nearest neighbors) and model-based algorithms (e.g., decision trees) have performed remarkably well in tackling the LR problem. Probabilistic Graphical Models (PGMs, e.g., Bayesian networks) have not been considered to deal with this problem because of the difficulty of modeling permutations in that framework. In this paper, we propose a Hidden Naive Bayes classifier (HNB) to cope with the LR problem. By introducing a hidden variable, we can design a hybrid Bayesian network in which several types of distributions can be combined: multinomial for discrete variables, Gaussian for numerical variables, and Mallows for permutations. We consider two kinds of probabilistic models: one based on a Naive Bayes graphical structure (where only univariate probability distributions are estimated for each state of the hidden variable) and another where we allow interactions among the predictive attributes (using a multivariate Gaussian distribution for the parameter estimation). The experimental evaluation shows that our proposals are competitive with the start-of-the-art algorithms in both accuracy and in CPU time requirements.

show abstract

Section: Hidden Naive Bayes For Label Rankingmentioning

confidence: 99%

Section: Hidden Naive Bayes For Label Rankingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Mixture-Based Probabilistic Graphical Models for the Label Ranking Problem

Rodrigo

Alfaro

Aledo

et al. 2021

Entropy

View full text Add to dashboard Cite

show abstract

“…Let us briefly mention some of these. The first group is formed by methods based on directly adapting well‐known machine learning algorithms, such as decision trees , 10 instance‐based learners , 10,22 probabilistic models , 22,23 association rules , 24 and neural networks ( multilayer perceptron ) 25 . The second group includes methods based on the transformation of the whole problem into a set of single‐class classifiers (e.g., labelwise , 26,27 pairwise approaches, 28,29 or chain classifiers 30 ).…”

Section: Preliminariesmentioning

confidence: 99%

Learning decision trees for the partial label ranking problem

2020

View full text Add to dashboard Cite

Ensemble learning for the partial label ranking problem

Alfaro

Aledo

Gámez

2022

Math Methods in App Sciences

View full text Add to dashboard Cite

The partial label ranking problem is a general interpretation of the preference learning scenario known as the label ranking problem, the goal of which is to learn preference classifiers able to predict a complete ranking with ties over the finite set of labels of the class variable. In this paper, we use unsupervised discretization techniques (equal-frequency and equal-width binning) to heuristically select the threshold for the numerical features in the algorithms based on induction of decision trees (partial label ranking trees algorithm). Moreover, we adapt the most well-known averaging (bootstrap aggregating and random forests) and boosting (adaptive boosting) ensemble methods to the partial label ranking problem, in order to improve the robustness of the built classifiers. We compare the proposed methods with the nearest neighbors-based algorithm (instance based partial label ranking) over the standard benchmark datasets, showing that our versions of the ensemble methods are superior in terms of accuracy. Furthermore, they are affordable in terms of computational efficiency.

show abstract

A Probabilistic Graphical Model-Based Approach for the Label Ranking Problem

Cited by 5 publications

References 18 publications

Mixture-Based Probabilistic Graphical Models for the Label Ranking Problem

Mixture-Based Probabilistic Graphical Models for the Label Ranking Problem

Learning decision trees for the partial label ranking problem

Ensemble learning for the partial label ranking problem

Contact Info

Product

Resources

About