ABSTRACTThe theory of rough sets is an extension set with two additional unary set-theoretic operators defined based on a binary relation the universe. These are related to modal in logics. By exploring relationship between and logics, this paper proposes examines number extended models. properties satisfied by relation, such as serial, reflexive, symmetric, transitive, Euclidean, various classes algebraic models can be derived. They correspond different logic systems. With respect graded...

In Brewer and Nash's Chinese Wall security policy model, there is a very strong implicit assumption that the 'conflict of interest' an equivalence relation. It called BN-axiom. Such axiom valid only for some special circumstances. By modifying their formulation, modified Nash model without BN-axiom defined. turns out to be rather 'conservative' in sense Walls are built within territory; it not really model. Next, new formal introduced which right on boundary China-an aggressive policy.< <ETX...

ABSTRACT The optimization of diverse industrial edge computing tasks presents a significant challenge due to the dynamic and heterogeneous nature operational demands. While deep reinforcement learning (DRL) has shown promise, task‐specific DRL models are often required in complex networks, such as real‐time anomaly detection latency‐sensitive decision‐making, increasing computational overhead. This leads large overheads, unstable performance, increased energy consumption. Such cost become...

Abstract The prediction of degradation and remaining useful life (RUL) for mechanical equipment has received widespread attention. Due to the large number components varying operating conditions, it is difffcult extract features. This paper proposes a generalized enhanced spatiotemporal global (ESTG) competitive network predicting life. model mainly includes spatial competition (SC) module, temporal (TC) (GC) module. SC module focuses on feature mining multi-source data, which can...

The notion of granular computing is examined. Obvious examples, such as fuzzy numbers, infinitesimal number and access control model, (pre-) topological spaces are A general models proposed; localized multi-level granulation can be modeled by generalized spaces, called neighborhood systems. For most Tarski type relational structures.

Mining association rules from databases is a time-consuming process. Finding the large item set fast crucial step in rule algorithm. In this paper we present algorithm (Bit-AssoRule) based on granular computing. Our Bit-AssocRule doesn't follow generation-and-test strategy of Apriori and adopts divide-and-conquer strategy, thus avoids table scan to rind prune itemsets, all operations finding itemsets datasets are bit its corresponding granular. The experimental result our with Apriori,...

The notion of fuzzy is context dependent, so for each very often there a theory. Present papers use the neighborhood systems to unify them. A system an association that assigns datum list data (a neighborhood). Rough sets and topological spaces are special cases. "real world" set should allow small amount perturbation, it have elastic membership function. Mathematically, such function can be expressed by highly structured subset space. Structured singletons, equivalence classes,...

In logic, a "real world" is modeled by Cantor set with relational structure. this paper, the structure confined to simplest kind, namely, binary relations. From different consideration, in granular computing, such has been called crisp/fuzzy granulation, or neighborhood system (FBNS). Intuitively, granulated into neighborhoods (generalized equivalence classes). Combining two views, kind of model BNS-space. view, classical theory knowledge representation universe whose finite relations;...

Deep graph prompt tuning (DeepGPT), which only tunes a set of continuous prompts for transformers, significantly decreases the storage usage during training. However, DeepGPT is limited by its uniform to input graphs with various structures. This because different structures dictate feature interactions between nodes, while are not dynamic tailor transformation topology. In this paper, we propose <u>T</u>opo-specific <u>G</u>raph <u>P</u>rompt <u>T</u>uning (TGPT ), provides topo-specific...

It is well known that the Kripke model for modal logic system S5 can be interpreted as an approximation space in rough set theory. In this paper, we generalize interpretation to relational granulation. We consider two multimodal logics reasoning about granulation open world and closed environments respectively. environment, objects are granulated into same equivalence class only if they have relationship with other objects, while a environment; objects. Such relations represented by derived...

In 1989, Brewer and Nash presented a Chinese Wall Security Policy model (CWSP); the exited whole conference. The intuition is to place on boundary of conflict interests classes so that information flow among competing companies cannot occur. Unfortunately, later in same year we pointed out their erroneous assumption relation (CIR) an equivalence relation; aggressive (ACWSP) was based analysis CIR as binary relation. Based our relative recent observation each induces relation, show ACWSP can...

Rough Set theory assumes the underlying structure of knowledge is a partition; each equivalence class represents piece basic knowledge. In this paper, we covering From which theories learning, approximations and reduction are developed.

Rough support vector machines (RSVMs) supplement conventional (SVMs) by providing a better representation of the boundary region. Increasing interest has been paid to theoretical development RSVMs, which already lead modification existing SVM implementations as RSVMs. This paper shows how extend use precision and recall from implementation RSVM implementation. Our approach is demonstrated in practice with help Gist, popular

The aim of this paper is to design an efficient and fast immune evolution algorithm for solving path planning pro- gram soccer robot. idea accurately read the inherent drawbacks existing algorithms (IA) propose new techniques resolve them. basic features IA dealt in are: hypermutation mechanism, clone expansion, memory several other fea- tures related initialization selection candidate so- lution present a population set. Dealing with above- mentioned we have proposed improved (IEA)....

Fuzzy set theory on a universe of granules (crisp sets) are considered. In this setting, the grade membership can be interpreted in most elementary fashion. A 1/2 means exactly one half granule (in suitable measure) belongs to fuzzy set. Based such views, theoretical operations very naturally induced from classical operations; they called R-operations. Some traditional operations, as max and min, R-operations specific classes sets. The granular view sets provides new healthy insight into how...