Conflict is widespread in society, and trisecting an agent set is a crucial research direction in three-way conflict analysis. In practice, the varying risk preferences among decision-makers lead to diverse trisections of an agent set in similar circumstances. In this paper, we consider decision-makers’ risk preferences and propose novel models of three-way conflict. Initially, we divide a set of issues into two disjoint subsets and utilize information entropy to compute issues’ weights. Then, we design alliance measures regarding an individual issue through the transition probability function. Based on the issues’ weights and alliance measures, we propose alliance probabilities between two agents. Additionally, we present risk-preferred, risk-averse, and risk-neutral decisions with the relative utility function. Finally, we design an algorithm to derive rules for three-way conflict analysis and demonstrate how to utilize the proposed model for decision-making through an example.
Decision trees (DTRs) and decision rules are extensively examined and applied in various domains of computer science. The theory of DTRs and rules highlights several crucial inquiries, such as how the complexity of deterministic decision trees (DDTRs) and decision rule systems depends on the complexity of the set of attributes associated with the columns of the decision table (DT). In this research paper, we focus on the analysis of nondeterministic decision trees (NDTRs) as a substitute for decision rule systems. NDTRs can be seen as representations of decision rule systems. We examine classes of DTs featuring multi-valued decisions (known as multi-label DTs) that maintain closure under attribute removal (columns) and modifications to the sets of assigned decisions for rows. We examine the behavior of functions that describe the worst-case dependence of the minimum complexity of DDTRs and NDTRs on the complexity of the set of attributes associated with the columns of tables belonging to a closed class (CC) of DTs closed under the above two operations. We enumerate all possible types of behavior exhibited by these functions.
Hyperbolic embedding has advantages for hierarchically category images, hence it has been applied in few-shot learning and achieved significant results. However, treating all samples equally may not ensure that the learned hyperbolic embedding adequately considers various modalities within the same category. To address this issue, this paper proposes a novel hyperbolic attention mechanism for few-shot learning, which adjusts weights based on the hyperbolic distance of samples to the average position. This approach balances the typicality and diversity of labeled samples, aiding the model in a deeper understanding of data structures. After hyperbolic embedding, weights were redistributed through this attention mechanism for few-shot learning. Experiments were conducted on the CUB and miniImageNet datasets. The experiments demonstrate the superiority of the proposed attention mechanism when using hyperbolic embedding for hierarchically category data.
In real-world scenarios, datasets characterized by nonlinear separability and fuzziness are quite common. Although fuzzy granular-balls generated by fuzzy c-means can capture the fuzziness of the datasets, it exhibits limitations in effectively handling nonlinearly separable datasets. Due to the advantage of spectral clustering in handling nonlinearly separable datasets, we fully leverage this capability to propose a novel clustering model, namely, fuzzy granular-balls based spectral clustering (FGBSC). It enhances spectral clustering by introducing fuzzy granular-balls as inputs, and effectively addresses the nonlinear separability of the datasets on the basis of capturing the fuzziness of the datasets. We perform experiments to evaluate the effectiveness of the proposed method in handling nonlinearly separable datasets.
Rough concepts have been introduced in [7] in the context of a mathematical framework unifying Rough Set Theory (RST) and Formal Concept Analysis (FCA). Algebraically, the lower and upper approximation operators on a concept lattice have similar order-theoretic properties to the and operators in modal logic. Thus, the logic of rough concepts has been defined as a (non-distributive) lattice-based modal logic whose relational semantics consists of formal contexts enriched with relations (interpreting the modal operators) satisfying the axioms classically corresponding to the reflexivity, symmetry, and transitivity of the accessibility relations of Kripke frames. Recently, the description logic LE- 𝒜ℒ𝒞 was introduced for reasoning in the semantic environment of these enriched formal contexts, and a tableaux algorithm was developed for checking the consistency of knowledge bases with acyclic TBoxes [5]. In the present paper, we introduce the description logic of rough concepts LE- 𝒜ℒ𝒞ℛ , which extends LE- 𝒜ℒ𝒞 with the (modal) axioms classically corresponding to reflexivity, symmetry, and transitivity, and develop its corresponding tableaux algorithm. We then introduce two extensions of LE- 𝒜ℒ𝒞ℛ : the first one (LE- 𝒜ℒ𝒞ℛ𝒪 ) extending LE- 𝒜ℒ𝒞ℛ with generated concepts, and the second one (LE- 𝒜ℒ𝒞ℛ𝒩 ) extending LE- 𝒜ℒ𝒞ℛ with feature-pair inconsistencies. The resulting description logic is a framework for modeling reasoning problems related to rough concepts, which is demonstrated through the case-study of a knowledge base for Whittaker’s five kingdom classification of living things.
This paper examines specific classes of conventional decision tables (DTs) that are closed under operations of attribute (column) removal and decision modifications assigned to rows. For DTs belonging to any of these closed classes (CCs), we investigate a greedy algorithm that constructs a deterministic decision tree. We demonstrate that the number of steps performed by this algorithm is limited by twice the number of rows in the table. Furthermore, we compare the behavior of two functions. The first function describes the increase in the minimum complexity of a deterministic decision tree for a DT from the CC in the worst-case scenario, in relation to the complexity of the set of attributes associated with columns of the table. The second function characterizes the worst-case complexity growth of the deterministic decision tree constructed by the greedy algorithm for a DT from the CC, considering the growth of complexity of the set of attributes associated with columns of the table. We divide the entire collection of pairs consisting of a bounded complexity measure (BCM) and a CC into three subsets. For each subset, we establish lower and upper bounds for the second function based on the first function.
This paper is an attempt to develop a notion of interactive information system, which contrary to the classical notion of information system allows real physical interactions to play a role in the process of gathering information about the objects or situations. The proposed notion of interactive information system is grounded on the basic building blocks of Interactive Granular Computing (IGrC), known as complex granules (c-granules). The main point of departure of IGrC from other existing mathematical tools for modeling complex real physical phenomenon is to include a possibility of introducing a process of learning and perceiving through interactions initiated in the real physical world. Thus, the model is not completely restricted in a pure mathematical manifold. It has both the abstract module and implementational module connected by a notion of control.
Functional dependency is an essential concept between attributes in relational algebra and tabular data analysis, and it is applied to the decomposition of tabular data. On the other hand, data dependency is a concept between attribute values. Functional dependency is usually given, and data dependency is often recognized after data analysis. We may recognize the hidden functional dependency as a particular case of data dependency. In this paper, we apply the rules obtained by the NIS-Apriori-based rule generator, which was implemented to handle rules from tabular and tabular data with missing values. We detect some candidates CONs of condition attributes that affect the decision attribute Dec using the obtained rules. We then apply the same rule generator specifying the detected one CON to determine the actual degree of dependency. This step eliminates the need to enumerate all CONs to understand dependencies and can handle extended dependencies for DIS and NIS. A running example using the implemented tools is also provided.
This research paper presents a novel approach to Rental Market Analysis for Property Management Firms using Large Language Models (LLMs) and Machine Learning techniques. The proposed system leverages LLM-based web scraping to extract data from dynamic websites, enabling the automated collection of relevant market information. By employing LLMs, the system generates insightful comparisons between property management firms and their listed properties, providing a comprehensive understanding of the competitive landscape. Additionally, an ensembled machine learning approach, utilizing multiple models, is developed to accurately predict rental prices. The integration of these cutting-edge technologies empowers property management firms with a dashboard that offers insightful analytics, predictive capabilities, and generated insights for data-driven decision-making. The system's architecture combines Python, ReactJS, AWS, PowerBI, PostgreSQL and OpenAI APIs to create a user-friendly interface that facilitates seamless data interaction and enhances insight generation. By automating data collection, analysis, and insight generation, this novel approach revolutionizes traditional rentalmarket analysis processes, enabling property management firms to stay competitive and optimize their business strategies in dynamic rental markets.
This paper presents a new approach to analyzing numerical data sets using covering-based rough sets based on the Mapper algorithm, a fundamental tool of topological data analysis (TDA). Specifically, by varying the parameters of Mapper, our approach generates different coverings from a numerical dataset that can be used to define lower approximations, and the associated quality of classification, for covering-based rough sets. We discuss the fundamental ideas of how to integrate both theories, and explore possible lines of further work.
The concept lattice plays a fundamental role in formal concept analysis (FCA) and finds widespread application across various fields. However, the presence of a large number of nodes in the concept lattice can pose challenges when it comes to comprehending the acquired conceptual knowledge. The size of the concept lattice is a significant concern in FCA, and obtaining an appropriately sized lattice is of utmost importance. To address this issue, this paper introduces a novel model for identifying important concepts in the concept lattice. The proposed model leverages concept indices and complex network analysis techniques to reduce the size of the lattice and enhance the understanding of conceptual knowledge. To derive the most valuable concepts, concept indices are first proposed by both node attribute information and structural information. Second, to fuse the attribute and structural information of concepts, an information system for concept indices is developed. In addition, the K-means method is employed to comprehensively evaluate all concept indices and obtain important concept identification results. Finally, an empirical study and comparative analyze demonstrate that the proposed model can effectively identify important concepts in the concept lattice.
We investigate approximation operators determined by arbitrary families of granules which cover the space of objects and without posing any conditions on the nature of granules. We recall from literature some rough sets approximation operators: classical equivalential rough sets operators by Z. Pawlak and their generalizations for tolerance relations: tolerance approximation operators by A. Skowron and J. Stepaniuk and tolerance-granular approximation operators suggested by Z. Pawlak. Then we proposed granular generalization of tolerance approaches to tolerance-based rough sets by means of arbitrary covering of the object space proposed by Y.Y. Yao Then we propose a version of tolerance-granular operators suggested by Z. Pawlak with biting procedure proposed in our paper. We prove basic properties of recalled tolerance rough sets operators. We discuss which pair of operators possesses the property of mutual definability. Then we show generalizations of tolerance rough sets approximation operators which possess mutual definability property by granular covering operators from.
In the past two years, large language models (LLMs) have shown extensive attention in the applications of intelligent transportation systems (ITS). Despite the huge potential, there is still a lack of comprehensive understanding of the advantages, challenges, and future efforts of LLMs in the transportation field. In this paper, we present a systematic investigation in this field, underlining their approaches and performance in improving forecasting accuracy, decision-making capability, and sim-to-real tasks. We first explore the current applications of LLMs in traffic management, transportation safety, and autonomous driving, as well as analyze their advantages and limitations. Then we also list some typical datasets employed within this domain. Challenges and prospects of the development of LLMs for ITS applications are discussed, encompassing technological, security, and policy aspects. We aim to offer a holistic overview of the transformative impact of LLMs in the transportation field, highlight their significance, and provide some possible views for future research and development.
Liver disease accounts for 4% of all deaths worldwide. Existing research on liver disease with rough set mostly studied from the perspective of clinical diagnosis, but has not yet been studied from the perspective of clinical medical test decision-making. To address the problem, this paper first converts the clinical diagnosis problem of liver disease into the clinical medical test decision-making problem of liver disease. Then the rough set variant of granular-ball rough set can be used to reduce the attributes of clinical medical test decision-making tasks of liver disease. Next, the clinical medical test decision-making problem is simplified to a classification task. Finally, the proposed method is experimentally verified on the processed liver disease dataset.
Considering the growing demands for efficient information retrieval from house rental markets by non-professional users, we develop a comprehensive framework for house information management, visualization, and prediction based on the CatBoost algorithm. We aim to promote the digital transformation of house rental market management and drive innovation in management methods. The conception and ideas of the Housing Rental Information Management and Prediction System are initially proposed, with subsequent application in Halifax, Canada. Integrating the Tableau server, database, and prediction model, we build a seamless web system to harmonize management, visualization, and prediction functionalities for rental house data. The details and effects of the application of the CatBoost algorithm within this system are emphasized, highlighting its precision, adaptability, and business viability in forecasting the house rental market.
Tracking moving targets in satellite videos has lately gained popularity. However, the growth of target tracking in satellite videos is slower than in general videos due to the following factors. Satellite video tracking faces challenges stemming from low frame rates, causing significant object movement between frames and impacting prediction accuracy, while high-resolution footage exacerbates tracking difficulties by requiring extensive search regions for targets occupying a minimal percentage of the total pixel count. In overall, the level of uncertainty surrounding the target in the satellite footage is excessive. To address the above problems, we propose a novel DiMP-based tracker and introduce SES to stabilize the estimation of target motion in satellite videos. Furthermore, we introduce an uncertainty measure, which is included into network outputs and a loss function to remove unreliable samples from the training set. Extensive experiments have demonstrated that our technique can track targets with the highest precision score and success rate on both the SatSOT and SV248S datasets, achieving state-of-the-art status.
Conflict often arises when individuals have different opinions. Owing to the ubiquity of conflict, conflict analysis is always widely discussed. Recently, the three-way conflict analysis proposed by Yao has attracted much attention. In Yao's framework, each three-way conflict model consists of the whole, trisections, and final results. However, the final results are induced once a trisection is given. In other words, the model cannot correct the final results. Therefore, we provide a new conflict analysis framework with negative feedback. This way, we can deliver better results, even when we are not given proper thresholds in advance. In addition, we provide three algorithms for three-way conflict models with negative feedback in this paper after showing the new framework. In the third algorithm, we focus on coalitions instead of trisections while improving thresholds, which distinguishes greatly from previous models.
The triangular fuzzy number intuitionistic fuzzy set, as an extension and generalization of intuitionistic fuzzy set, has more advantages than the single value representation of membership degree and non-membership degree and interval number representation of intuitionistic fuzzy set. In this paper, the triangular fuzzy number intuitionistic fuzzy set and rough set are fused, and the triangular fuzzy number intuitionistic fuzzy rough set model is constructed based on the intuitionistic fuzzy approximation relation of triangle fuzzy number. Firstly, the approximate operators, correlation properties and three-way regions are discussed and verified by numerical examples. Furthermore, combined with the covering rough set, the triangular fuzzy number intuitionistic covering rough set is constructed, which can be used for multi-attribute decision making. Then, combined with the covering rough set, the triangular fuzzy number intuitionistic fuzzy covering rough set is constructed, which can be used for multi-attribute decision making. Finally, the multi-attribute decision making is carried out with an example considering the expert weight. The results show that the triangular fuzzy number intuitionistic fuzzy covering rough set become novel and effective for multi-attribute decision making. This study provides an in-depth insight into decision making from model and method.
The article presents the RIONIDA learning algorithm based on combination of two widely-used empirical approaches: rule induction and instance-based learning for imbalanced data classification. The algorithm is a substantial extension of the well-known RIONA algorithm developed for balanced data. RIONIDA is relatively fast and significantly outperforms the state-of-the-art algorithms analysed in the paper.
In the current extended variable precision rough set models (VPRS models), a granule is a set of objects. Examples of such models include the variable precision neighbourhood rough set model and the variable precision covering-based rough set model. These objects can be considered as real vectors when dealing with real-valued data, and, therefore, a granule is effectively a set of real vectors. In this paper, we show that a single vector can be considered as a granule, too. More precisely, we introduce the granule vector and the approximation vector, and show that the negativity of the inner product of these two vectors is equivalent to a granule being contained the lower (or upper) approximation. Based on this equivalence, we propose a novel extension of VPRS model which treats a single vector as a granule. In particular, this generalised model can deal with applications not covered by the existing models.
