The granular extension of Sugeno-type fuzzy models based on optimal allocation of information granularity and its application to forecasting of time series

The Sugeno-type fuzzy models are used frequently in system modeling. The idea of information granulation inherently arises in the design process of Sugeno-type fuzzy model, whereas information granulation is closely related with the developed information granules. In this paper, the design method of Sugeno-type granular model is proposed on a basis of an optimal allocation of information granularity. The overall design process initiates with a well-established Sugeno-type numeric fuzzy model (the original Sugeno-type model). Through assigning soundly information granularity to the related parameters of the antecedents and the conclusions of fuzzy rules of the original Sugeno-type model (i.e. granulate these parameters in the way of optimal allocation of information granularity becomes realized), the original Sugeno-type model is extended to its granular counterpart (granular model). Several protocols of optimal allocation of information granularity are also discussed. The obtained granular model is applied to forecast three real-world time series. The experimental results show that the method of designing Sugeno-type granular model offers some advantages yielding models of good prediction capabilities. Furthermore, those also show merits of the Sugeno-type granular model: (1) the output of the model is an information granule (interval granule) rather than the specific numeric entity, which facilitates further interpretation; (2) the model can provide much more flexibility than the original Sugeno-type model; (3) the constructing approach of the model is of general nature as it could be applied to various fuzzy models and realized by invoking different formalisms of information granules.     

MGRS: A multi-granulation rough set

The original rough set model was developed by Pawlak, which is mainly concerned with the approximation of sets described by a single binary relation on the universe. In the view of granular computing, the classical rough set theory is established through a single granulation. This paper extends Pawlak’s rough set model to a multi-granulation rough set model (MGRS), where the set approximations are defined by using multi equivalence relations on the universe. A number of important properties of MGRS are obtained. It is shown that some of the properties of Pawlak’s rough set theory are special instances of those of MGRS.Moreover, several important measures, such as accuracy measureα, quality of approximationγ and precision of approximationπ, are presented, which are re-interpreted in terms of a classic measure based on sets, the Marczewski-Steinhaus metric and the inclusion degree measure. A concept of approximation reduct is introduced to describe the smallest attribute subset that preserves the lower approximation and upper approximation of all decision classes in MGRS as well. Finally, we discuss how to extract decision rules using MGRS. Unlike the decision rules (“AND” rules) from Pawlak’s rough set model, the form of decision rules in MGRS is “OR”. Several pivotal algorithms are also designed, which are helpful for applying this theory to practical issues. The multi-granulation rough set model provides an effective approach for problem solving in the context of multi granulations.     

Building the fundamentals of granular computing: A principle of justifiable granularity

The study introduces and discusses a principle of justifiable granularity, which supports a coherent way of designing information granules in presence of experimental evidence (either of numerical or granular character). The term “justifiable” pertains to the construction of the information granule, which is formed in such a way that it is (a) highly legitimate (justified) in light of the experimental evidence, and (b) specific enough meaning it comes with a well-articulated semantics (meaning). The design process associates with a well-defined optimization problem with the two requirements of experimental justification and specificity. A series of experiments is provided as well as a number of constructs carried for various formalisms of information granules (intervals, fuzzy sets, rough sets, and shadowed sets) are discussed as well.     

Granular worlds: Representation and communication problems

In this study, we introduce a concept of granular worlds and elaborate on various representation and communication issues arising therein. A granular world embodies a collection of information granules being regarded as generic conceptual entities used to represent knowledge and handle problem solving. Granular computing is a paradigm supporting knowledge representation, coping with complexity, and facilitating interpretation of processing. In this sense, it is crucial to all man‐machine pursuits and data mining and intelligent data analysis, in particular. There are two essential facets that are inherently associated with any granular world, that is a formalism used to describe and manipulate information granules and the granularity of the granules themselves (roughly speaking, by the granularity we mean a “size” of such information granules; its detailed definition depends upon the formal setting of the granular world). There are numerous formal models of granular worlds ranging from set‐theoretic developments (including sets, fuzzy sets, and rough sets) to probabilistic counterparts (random sets, random variables and alike). In light of the evident diversity of granular world (occurring both in terms of the underlying formal settings as well as levels of granularity), we elaborate on their possible interaction and identify implications of such communication. More specifically, we have cast these in the form of the interoperability problem that is associated with the representation of information granules. © 2000 John Wiley & Sons, Inc.     

多粒度决策系统属性约简的最优粒度选择

粒计算理论从多个角度、多个不同的粒度层次出发,对不确定、不精确或复杂的问题进行求解,现已成为人工智能领域研究的一种重要方法。针对决策系统属性约简与高效决策的粒度选择问题,分析了多粒度决策系统中信息粒与粒度划分的概念,定义了粒化度量和粒结构关于对象的粒化粗糙度,能够准确地反映决策系统中不同粒结构下的知识粒度大小。为弥补传统决策系统约简往往只考虑基于论域属性约简的缺陷,讨论了基于对象的局部约简方法,提出了基于论域和对象的决策系统最优粒度选择约简算法,并结合实例验证了该算法的有效性。     

A design of granular fuzzy classifier

In this paper, we propose a new design methodology of granular fuzzy classifiers based on a concept of information granularity and information granules. The classifier uses the mechanism of information granulation with the aid of which the entire input space is split into a collection of subspaces. When designing the proposed fuzzy classifier, these information granules are constructed in a way they are made reflective of the geometry of patterns belonging to individual classes. Although the elements involved in the generated information granules (clusters) seem to be homogeneous with respect to the distribution of patterns in the input (feature) space, they still could exhibit a significant level of heterogeneity when it comes to the class distribution within the individual clusters. To build an efficient classifier, we improve the class homogeneity of the originally constructed information granules (by adjusting the prototypes of the clusters) and use a weighting scheme as an aggregation mechanism.     

Comments on “Generalised finite Radon transform for N × N images”

In Kingston and Svalbe [1], a generalized finite Radon transform (FRT) that applied to square arrays of arbitrary size N × N was defined and the Fourier slice theorem was established for the FRT. Kingston and Svalbe asserted that “the original definition by Matúš and Flusser was restricted to apply only to square arrays of prime size,” and “Hsung, Lun and Siu developed an FRT that also applied to dyadic square arrays,” and “Kingston further extended this to define an FRT that applies to prime-adic arrays”. It should be said that the presented generalized FRT together with the above FRT definitions repeated the known concept of tensor representation, or tensor transform of images of size N × N which was published earlier by Artyom Grigoryan in 1984-1991 in the USSR. The above mentioned “Fourier slice theorem” repeated the known tensor transform-based algorithm of 2-D DFT [5-11], which was developed for any order N₁ × N₂ of the transformation, including the cases of N × N, when N = 2^r, (r > 1), and N = L^r, (r ≥ 1), where L is an odd prime. The problem of “over-representation” of the two-dimensional discrete Fourier transform in tensor representation was also solved by means of the paired representation in Grigoryan [6-9].     

作物产量预测的粒度分析法

针对作物产量预测,提出基于商空间粒度计算的分析法。在商空间粒度计算理论思想下,分析作物产量序列中粒度的选取,用属性划分方法对论域X进行颗粒化,对属性f取不同的粒度进行颗粒化。通过属性的粒度变化对论域进行划分,得到新的商空间并应用其解决问题,可以降低问题复杂度。通过商空间理论中的分层与合成技术选取大小合适的粒度,能全面获取产量序列中的信息,也更加符合人类智能特点。冬小麦产量预测实验结果也证明这种粒度分析和选取方法是有效的。     

From local neural networks to granular neural networks: A study in information granulation

In this study, we are concerned with a construction of granular neural networks (GNNs)—architectures formed as a direct result reconciliation of results produced by a collection of local neural networks constructed on a basis of individual data sets. Being cognizant of the diversity of the results produced by the collection of networks, we arrive at the concept of granular neural network, producing results in the form of information granules (rather than plain numeric entities) that become reflective of the diversity of the results generated by the contributing networks. The design of a granular neural network exploits the concept of justifiable granularity. Introduced is a performance index quantifying the quality of information granules generated by the granular neural network. This study is illustrated with the aid of machine learning data sets. The experimental results provide a detailed insight into the developed granular neural networks.     

基于局部广义多粒度粗糙集的多标记最优粒度选择

在多粒度粗糙集模型中,粒度选择总是与正域有关.由于全体标记确定对象集上的分类过细,落入正域的对象很少或为空集,导致正域约简方法可能丢失大量信息甚至失效.为了克服这一缺陷,文中提出基于局部广义多粒度粗糙集的多标记最优粒度选择方法.首先,引入广义局部多粒度粗糙集的相关概念,通过设置信息水平参数,对单个标记的对象集合进行近似.然后,通过定义多粒度多标记信息系统的粒度质量,给出粒度重要性.最后,设计最优粒度选择的启发式算法,并通过实例验证文中方法的有效性     

A novel method for constructing the optimal hierarchical structure based on fuzzy granular space

Granular computing serves as a general framework for complex problem solving in broad scopes and at various levels. The granularity was constructed via many ways, however, for complex systems there remain two challenges including determining a reasonable granularity and extracting the hierarchical information. In this paper, a new method is presented for constructing the optimal hierarchical structure based on fuzzy granular space. Firstly, the inter-class deviations and intra-class deviations were introduced, whose properties were investigated in depth and approved mathematically. Secondly, the fuzzy hierarchical evaluation index is developed, followed with a novel model for extracting the global optimal hierarchical structure established. An algorithm is then proposed, which reliably constructs the multi-level structure of complex system. Finally, to reduce the complexity, the granular signatures are extracted according to the nearest-to-center principle; with the use of the signatures, a classifier is designed for verifying our method. The validation of this method is approved by an application to the H1N1 influenza virus system. The theories and methodologies on granular computing presented here are helpful for capturing the structural information of complex system, especially for data mining and knowledge discovery.     

Theory and applications of granular labelled partitions in multi-scale decision tables

Granular computing and acquisition of if-then rules are two basic issues in knowledge representation and data mining. A formal approach to granular computing with multi-scale data measured at different levels of granulations is proposed in this paper. The concept of labelled blocks determined by a surjective function is first introduced. Lower and upper label-block approximations of sets are then defined. Multi-scale granular labelled partitions and multi-scale decision granular labelled partitions as well as their derived rough set approximations are further formulated to analyze hierarchically structured data. Finally, the concept of multi-scale information tables in the context of rough set is proposed and the unravelling of decision rules at different scales in multi-scale decision tables is discussed.     

基于双论域笛卡尔积的粗糙熵与知识粒度

基于笛卡尔积,确立双直积论域覆盖空间,并研究其中的粗糙熵与知识粒度.首先,将双论域近似空间诱导出两个单论域覆盖空间,构建双直积论域覆盖空间.将双论域粗糙熵与知识粒度定位于一个单论域覆盖空间.通过结构模拟与粒替换,确定对称单论域覆盖空间与双直积论域覆盖空间的粗糙熵与知识粒度.对于三套双度量,得到相关的双量和、上下确界、粒化单调性及三支线性组合性.最后,通过数据模拟与仿真实验验证度量构建与理论性质的有效性.     

Granular neural networks: A study of optimizing allocation of information granularity in input space

In this paper, we develop a granular input space for neural networks, especially for multilayer perceptrons (MLPs). Unlike conventional neural networks, a neural network with granular input is an augmented study on a basis of a well learned numeric neural network. We explore an efficient way of forming granular input variables so that the corresponding granular outputs of the neural network achieve the highest values of the criteria of specificity (and support). When we augment neural networks through distributing information granularities across input variables, the output of a network has different levels of sensitivity on different input variables. Capturing the relationship between input variables and output result becomes of a great help for mining knowledge from the data. And in this way, important features of the data can be easily found. As an essential design asset, information granules are considered in this construct. The quantification of information granules is viewed as levels of granularity which is given by the expert. The detailed optimization procedure of allocation of information granularity is realized by an improved partheno genetic algorithm (IPGA). The proposed algorithm is testified effective by some numeric studies completed for synthetic data and data coming from the machine learning and StatLib repositories. Moreover, the experimental studies offer a deep insight into the specificity of input features.     

基于粒计算的Apriori算法及其在图书管理系统中的应用

粒计算作为一种新的信息和知识处理的方法近来已经被许多研究者所重视,以及在许多领域中的得到应用。本质上,粒计算能够表示存储在系统中的数据的语义信息,因此粒计算能作为用于探索数据性质的一种方法,如挖掘数据库中的关联规则。本文在分析经典Apriori算法的基础上,从信息粒的角度出发,提出基于粒计算生成k-频繁项目集算法。分析了对给定问题,当用粒计算模型求解时需要解决的几个基本问题。最后通过实例说明如何通过信息粒的二进制表示,并基于粒计算k-频繁项目集生成算法来获取隐藏在图书借还记录中的有关关联规则。可以看出该算法具有实际应用价值。     

基于模糊知识粒度的混合属性约简算法

现实世界中常常包含着海量的、不完整的、模糊及不精确的数据或对象,使得模糊信息粒化成为近年来研究趋势。利用论域上的模糊等价关系定义了模糊粒度世界的模糊知识粒度,给出了新的属性约简条件和核属性计算方法,以便更好地挖掘出潜在的、有利用价值的信息。针对粗糙集在对连续属性约简的过程中容易造成信息缺失和不能对模糊属性处理的现象,提出了一种基于模糊知识粒度对混合决策系统约简的启发式算法,省去了连续属性离散化过程,减少了计算量,为离散值域和混合值域约简提供了统一的方法。最后通过实例验证了其有效性。     

A molecular dynamical study of granular fluids I: The unforced granular gas in two dimensions

Results of a numerical study of the dynamics of a collection of disks colliding inelastically in a periodic two-dimensional enclosure are presented. The properties of this system, which is perhaps the simplest model for rapidly flowing granular materials, are markedly different from those known for atomic or moleclar gases, in which collisions are of elastic nature. The most prominent feature characterizing granular systems, even in the idealized situation in which no external forcing exists and the initial condition is statistically homogeneous, is their inherent instability to inhomogeneous fluctuations. Granular gases are thus generically nonuniform, a fact that suggests extreme caution in pursuing direct analogies with molecular gases. We find that once an inhomogeneous state sets in, the velocity distribution functions differ from the classical Maxwell-Boltzmann distribution. Other characteristics of the system are different from their counterparts in molecular systems as well. For a given value of the coefficient of restitution,e, a granular system forms clusters of typical separationL ₀l/(1-e ²)^1/2, wherel is the mean free path in the corresponding homogeneous system. Most of the fluctuating kinetic energy then resides in the relatively dilute regions that surround the clusters. Systems whose linear dimensions are less thenL ⁰ do not give rise to clusters; still they are inhomogeneous, the scale of the corresponding inhomogeneity being the longest wavelength allowed by the system's size. The present article is devoted to a detailed numerical study of the above-mentioned clustering phenomenon in two dimensions and in the absence of external forcing. A theoretical framework explaining this phenomenon is outlined. Some general implications as well as practical ramifications are discussed.     

Living in the Hutt Space: Immersive process in the Star Wars Role-Play community of Second Life

Immersive virtual settings are evolving to become new “spaces of life”. Humans inhabit these different virtual worlds through their avatars, and tend to gather into communities. However, the behavioral factors underlying the cognitive process of immersion in virtual worlds are still far to be understood. We here investigated these factors using the Star Wars Role-Play community of the virtual setting of Second Life as a model. More specifically, our studies focused on the immersion process in the “Hutt Space”, a portion of the Star Wars Galaxy ruled by the alien species of the Hutts, which combines the trademark aspects of Star Wars universe. Using both quantitative and qualitative methods, we identified some of the factors which favor the immersion process. Our results suggest that the different behavioral factors contributing to the immersion process can be organized in three structuring dimensions: commitment, cohesion, and coherence. We also unveil a compensatory mechanism between appearance and behavioral factors in creation and maintenance of social groups in virtual worlds. Finally, we point out some of the behavioral aspects of the evolution from passive media engagement (spectators), to active media engagement (actors), and suggest a theoretical framework to investigate how human inhabit immersive virtual spaces.     

On correction equations and domain decomposition for computing invariant subspaces

By considering the eigenvalue problem as a system of nonlinear equations, it is possible to develop a number of solution schemes which are related to the Newton iteration. For example, to compute eigenvalues and eigenvectors of an n × n matrix A, the Davidson and the Jacobi-Davidson techniques, construct ‘good’ basis vectors by approximately solving a “correction equation” which provides a correction to be added to the current approximation of the sought eigenvector. That equation is a linear system with the residual r of the approximated eigenvector as right-hand side.One of the goals of this paper is to extend this general technique to the “block” situation, i.e., the case where a set of p approximate eigenpairs is available, in which case the residual r becomes an n × p matrix. As will be seen, solving the correction equation in block form requires solving a Sylvester system of equations. The paper will define two algorithms based on this approach. For symmetric real matrices, the first algorithm converges quadratically and the second cubically. A second goal of the paper is to consider the class of substructuring methods such as the component mode synthesis (CMS) and the automatic multi-level substructuring (AMLS) methods, and to view them from the angle of the block correction equation. In particular this viewpoint allows us to define an iterative version of well-known one-level substructuring algorithms (CMS or one-level AMLS). Experiments are reported to illustrate the convergence behavior of these methods.     

不协调多粒度标记决策系统最优粒度的对比*

针对具有多粒度标记的不协调决策系统的知识表示和知识获取问题展开研究.首先,介绍多粒度标记信息系统的概念,在多粒度标记信息系统中定义不可分辨关系.然后,给出由不同粒度层面下信息粒度的表示及其相互关系,并进一步定义在不同粒度层面下集合的下、上近似概念,并讨论它们性质.最后,介绍不协调多粒度标记决策系统中8种协调性和最优粒度概念,并讨论它们之间的相互关系.