Article
Computer Science, Artificial Intelligence
Xiaoyu Han, Xiubin Zhu, Witold Pedrycz, Zhiwu Li
Summary: This study designs a three-way classification mechanism by combining fuzzy decision trees and expressing uncertainty. A fuzzy decision tree is constructed through generalization and the three-way decision model is widely used. An efficient way to flag uncertain data is proposed, which is not possible with commonly used fuzzy decision trees. The developed mechanism consists of two stages: building a fuzzy decision tree and determining the uncertainty level to reject instances. The rejection quality is quantified in terms of accuracy and coefficient, and the mechanism performs better than other three-way decision models.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Shlomi Maliah, Guy Shani
Summary: Cost sensitive classification aims to minimize the expected cost by deciding on the next attribute to measure after each observation. This paper suggests using POMDPs for cost sensitive classification and identifies potentially important belief states through standard decision trees.
ARTIFICIAL INTELLIGENCE
(2021)
Article
Computer Science, Artificial Intelligence
Dominik Hose, Michael Hanss
Summary: The study introduces a novel Imprecise Probability-to-Possibility Transformation that unifies many results in quantitative possibility theory concerning information modeling, data analysis, and the construction of joint distributions. Furthermore, it demonstrates how it enables new results about possibilistic information aggregation and how it may refine frequentist inference in statistical models.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2021)
Article
Computer Science, Artificial Intelligence
Vinicius G. Costa, Carlos E. Pedreira
Summary: This paper reviews the recent advances in Decision Trees (DTs) research, focusing on issues related to fitting training data, generalization, and interpretability, as well as providing an overview of the field, its key concerns, and future trends.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Computer Science, Artificial Intelligence
Pouya Shati, Eldan Cohen, Sheila A. A. McIlraith
Summary: Decision trees are popular in machine learning for their interpretability and performance, but existing methods for constructing them have limitations. This study presents a SAT-based encoding approach that directly supports non-binary data in decision tree construction. Experimental results demonstrate that our approach outperforms existing methods on non-binary datasets and has lower memory consumption.
Article
Mathematics
Ghous Ali
Summary: This paper aims to develop a novel fuzzy parameterized possibility fuzzy bipolar soft set model to deal with complicated uncertainties in multicriteria decision-making problems. The proposed model efficiently describes the possibility of fuzzy belongingness of alternatives under fuzzy parameterized bipolar parameters and presents respective operations and properties. Furthermore, similarity measures between fuzzy parameterized possibility fuzzy bipolar soft sets are proposed and applied to an agricultural land selection scenario, showing the eminent quality of the proposed work over existing ones through comparative analysis.
JOURNAL OF MATHEMATICS
(2023)
Article
Physics, Multidisciplinary
Mikhail Moshkov
Summary: This paper investigates the application of decision trees in infinite binary information systems based on rough set theory, test theory, and exact learning. By defining the notion of a problem over an information system and studying three functions of the Shannon type, the authors explore the dependence between the depth of a decision tree and the number of attributes in the problem description. The obtained results classify the infinite binary information systems into four complexity classes.
Article
Computer Science, Artificial Intelligence
Maxime Amram, Jack Dunn, Ying Daisy Zhuo
Summary: This research proposes an approach for directly learning optimal tree-based prescription policies from data. It combines methods for counterfactual estimation from the causal inference literature with recent advancements in training globally-optimal decision trees. The resulting optimal policy trees demonstrate excellent performance across various problems, while also being interpretable and scalable.
Article
Computer Science, Information Systems
Andrea Campagner, Davide Ciucci, Carl-Magnus Svensson, Marc Thilo Figge, Federico Cabitza
Summary: This study examines the impact of Ground Truth quality on the performance of Machine Learning models, introducing three reduction concepts to handle this uncertainty and demonstrating that the proposed algorithms outperform state-of-the-art approaches in addressing this form of uncertainty.
INFORMATION SCIENCES
(2021)
Article
Physics, Multidisciplinary
Yuta Nakahara, Shota Saito, Akira Kamatsuka, Toshiyasu Matsushima
Summary: This paper discusses the recursive structure of full rooted trees and their applications in statistical models, proposing a method to assume a prior distribution on full rooted trees for model selection based on Bayes decision theory. It also describes a method for calculating the properties of a probability distribution, such as mode, expectation, and posterior distribution.
Article
Computer Science, Information Systems
Yangxue Li, Enrique Herrera-Viedma, Gang Kou, Juan Antonio Morente-Molinera
Summary: This paper proposes a Z-number-valued rule-based decision tree (ZRDT) and provides the learning algorithm. Compared with other classical decision trees, ZRDT performs better in terms of classification accuracy and decision tree size. ZRDT uses information gain to select features in each rule instead of fuzzy confidence, and generates a second fuzzy number with negative samples to improve the model's fit to the training data. Based on statistical tests, ZRDT achieves the highest classification performance with the smallest size for the produced decision tree.
INFORMATION SCIENCES
(2023)
Review
Computer Science, Information Systems
Leonardo Canete-Sifuentes, Raul Monroy, Miguel Angel Medina-Perez
Summary: Decision trees are popular due to their interpretability, Multivariate Decision Trees are used to improve classification performance, but there is a lack of adequate statistical comparison for understanding the capabilities of existing algorithms.
Article
Operations Research & Management Science
Oktay Gunluk, Jayant Kalagnanam, Minhan Li, Matt Menickelly, Katya Scheinberg
Summary: The paper introduces a mixed integer programming approach to construct optimal decision trees taking into account the special structure of categorical features and allowing combinatorial decisions at each node. The method can also handle numerical features and achieve very good accuracy with small training sets. The optimization problems solved are tractable with modern solvers.
JOURNAL OF GLOBAL OPTIMIZATION
(2021)
Article
Environmental Sciences
Jianmei Ling, Lu Li, Haiyan Wang
Summary: Compared with traditional optical and multispectral remote sensing images, hyperspectral images with hundreds of bands provide the possibility of fine classification of the earth's surface. The novel hyperspectral image classification method based on a segment forest (SF) effectively improves classification accuracy, enhancing overall accuracy by around 11-19%. The SF-based hyperspectral image classification improves accuracy and efficiency compared with other algorithms, proving its effectiveness.
Article
History & Philosophy Of Science
Hassan Amiriara
Summary: The discussion revolves around the relationship between the standard formulation of the Theory of Relativity and the philosophy of time, particularly regarding the implications for eternalism. While some argue that Relativity strongly supports eternalism as the only possible ontology of time, others propose new objections to this conclusion without denying the relevance of certain relations.
Article
Computer Science, Artificial Intelligence
Timotheus Kampik, Kristijonas Cyras, Jose Ruiz Alarcon
Summary: This paper presents a formal approach to explaining changes in inference in Quantitative Bipolar Argumentation Frameworks (QBAFs). The approach traces the causes of strength inconsistencies and provides explanations for them.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Xiangnan Zhou, Longchun Wang, Qingguo Li
Summary: This paper aims to establish a closer connection between domain theory and Formal Concept Analysis (FCA) by introducing the concept of an optimized concept for a formal context. With the utilization of optimized concepts, it is demonstrated that the class of formal contexts directly corresponds to algebraic domains. Additionally, two subclasses of formal contexts are identified to characterize algebraic L-domains and Scott domains. An application is presented to address the open problem of reconstructing bounded complete continuous domains using attribute continuous contexts, and the presentation of algebraic domains is extended to a categorical equivalence.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Sihan Wang, Zhong Yuan, Chuan Luo, Hongmei Chen, Dezhong Peng
Summary: Anomaly detection is widely used in various fields, but most current methods only work for specific data and ignore uncertain information such as fuzziness. This paper proposes an anomaly detection algorithm based on fuzzy rough entropy, which effectively addresses the similarity between high-dimensional objects using distance and correlation measures. The algorithm is compared and analyzed with mainstream anomaly detection algorithms on publicly available datasets, showing superior performance and flexibility.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Mario Alviano, Francesco Bartoli, Marco Botta, Roberto Esposito, Laura Giordano, Daniele Theseider Dupre
Summary: This paper investigates the relationships between a multipreferential semantics in defeasible reasoning and a multilayer neural network model. Weighted knowledge bases are considered for a simple description logic with typicality under a concept-wise multipreference semantics. The semantics is used to interpret MultiLayer Perceptrons (MLPs) preferentially. Model checking and entailment based approach are employed in verifying conditional properties of MLPs.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Bazin Alexandre, Galasso Jessie, Kahn Giacomo
Summary: Formal concept analysis is a mathematical framework that represents the information in binary object-attribute datasets using a lattice of formal concepts. It has been extended to handle more complex data types, such as relational data and n-ary relations. This paper presents a framework for polyadic relational concept analysis, which extends relational concept analysis to handle relational datasets consisting of n-ary relations.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Ander Gray, Marcelo Forets, Christian Schilling, Scott Ferson, Luis Benet
Summary: The presented method combines reachability analysis and probability bounds analysis to handle imprecisely known random variables. It can rigorously compute the temporal evolution of p-boxes and provide interval probabilities for formal verification problems. The method does not impose strict constraints on the input probability distribution or p-box and can handle multivariate p-boxes with a consonant approximation method.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Laszlo Csato
Summary: This paper studies a special type of incomplete pairwise comparison matrices and proposes a new method to determine the missing elements without violating the ordinal property.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)