Abstract:In contrast to multi-class classification, in multi-label classification (MLC), each data sample is a member of several classes. In this regard, the classifiers should discover more complex patterns within the data; accordingly, the complexity grows remarkably. Recently, scholars used the binary relevance, the label powerset, and the ensemble method (EM) techniques to extend state-of-the-art classifiers or develop new ones to solve real-world MLC problems more optimally. The current paper recruits the ridge regression (RR) and a decision tree (DT) and presents a new MLC model named PHMLC. In the PHMLC, the Laplacian Eigenmaps Nonlinear Dimensionality Reduction Algorithm is employed to omit the redundant and irrelevant features. Next, the data is encoded by the principal label space transformation (PLST) and singular value decomposition. Then, the RR is recalled to predict the first set of labels through the encoded data. Besides, the Binary Relevance technique is applied to the original DT, and an MLC model called BR-DT is introduced. The BR-DT is responsible for predicting the second set of labels. Eventually, the greedy selection mechanism is utilized to obtain final labels based on the predicted labels by the PLST-based RR and the BR-DT models. To demonstrate the PHMLC's effectiveness, the PHMLC model is applied to nine real-life multi-label datasets, and the results are compared with the BR-SVM, CPLST, BR-Ridge, BR-DR, PLST, and BR-DT models regarding the AP, EM, HS, Macro, and Micro F1 metrics.

Multi-valued attribute and multi-labeled data decision tree algorithm

Algorithm of Multi-valued Attribute and Multi-labeled Data Decision Tree

Multi-agent-system-based Multi-Class Transductive Learning

Novel Design of Decision-Tree-Based Support Vector Machines Multi-Class Classifier

Combine Multi-Valued Attribute Decomposition with Multi-Label Learning

Multi-path Decision Tree

Multi-view and Multi-label Method with Three-Way Decision-Based Clustering.

Heuristic Mode Research and Application of Decision Tree Algorithm

An Incremental Decision Tree for Mining Multilabel Data

A Multi-Classifier Combined Decision Tree Hierarchical Classification Method

Decision by Maximum of Posterior Probability Average with Weights: a Method of Multiple Classifiers Combination

An Improved Id3 Decision Tree Algorithm

An Optimized Multi-class Classification Algorithm Based on SVM Decision Tree

Decision Tree Support Vector Machine Based on Genetic Algorithm for Multi-Class Classification

A hybrid principal label space transformation-based ridge regression and decision tree for multi-label classification

Knowledge Acquisition Methods for Multi-Label Decision System Based on Rough Sets

Three-Way Decisions Based Multi-label Learning Algorithm with Label Dependency

Multi-label Classification Algorithm Based on Association Rules

Multi-label Emotion Classification Based on Decision-Theoretic Rough Set

Multi-Task Sparse Discriminant Analysis (Mtsda) With Overlapping Categories

ML-FOREST: A Multi-Label Tree Ensemble Method for Multi-Label Classification