Knowledge transfer in system modeling and its realization through an optimal allocation of information granularity

Fuzzy models occupy one of the dominant positions on the research agenda of fuzzy sets exhibiting a wealth of conceptual developments and algorithmic pursuits as well as a plethora of applications. Granular fuzzy modeling dwelling on the principles of fuzzy modeling opens new horizons of investigations and augments the existing design methodology exploited in fuzzy modeling. In a nutshell, granular fuzzy models are constructs built upon fuzzy models or a family of fuzzy models. We elaborate on a number of compelling reasons behind the emergence of granular fuzzy modelling, and granular modeling, in general. Information granularity present in such models plays an important role. Given a fuzzy model M, the associated granular model incorporates granular information to quantify a performance of the original model, facilitate collaborative pursuits of knowledge management and knowledge transfer. We discuss several main categories of granular fuzzy models where such categories depend upon the formalism of information granularity giving rise to interval-valued fuzzy models, fuzzy fuzzy model (fuzzy2 models, for short), and rough -fuzzy models. The design of granular fuzzy models builds upon two fundamental concepts of Granular Computing: the principle of justifiable granularity and an optimal allocation (distribution) of information granularity. The first one supports a construction of information granules of a granular fuzzy model. The second one emphasizes the role of information granularity being treated as an important design asset. The underlying performance indexes guiding the design of granular fuzzy models are discussed and a multiobjective nature of the construction of these models is stressed.Keywordsinformation granularitygranular fuzzy modelsprinciple of justifiable granularityknowledge managementgranularity allocation

Building a consensus for the best-worst method in group decision-making with an optimal allocation of information granularity

In this study, we propose a new concept of granular rule-based models whose rules assume a format ``if G(A i ) then G(f i )'' where G$(.)s are granular generalizations of the numeric conditions and conclusions of the rules. Those generalizations can be expressed e.g., in terms of interval-valued, type-2 or probabilistic fuzzy sets. We discuss several classes of fuzzy models depending upon available information granules and offer a motivation present behind their emergence. The design of these granular architectures exploits the essentials of Granular Computing such as a principle of justifiable granularity and an optimal allocation of information granularity. Detailed investigations of the performance indexes (objective functions) along with the related optimization schemes are covered as well.

Granular transfer learning

In this study, we pursue a new direction for system modeling by introducing the concept of granular models, which produce results in the form of information granules (such as intervals, fuzzy sets, and rough sets). We present a rationale and several key motivating arguments behind the use of granular models and discuss their underlying design processes. The development of the granular model includes optimal allocation of information granularity through optimizing the criteria of coverage and specificity. The emergence and construction of granular models of type-2 and type-n (in general) is discussed. It is shown that achieving a suitable coverage-specificity tradeoff (compromise) is essential for developing granular models.

Local sources of knowledge structured in the form of logic descriptors—constructs of fuzzy logic, are arranged together (structured) in the form of a global model coming as a high-level granular logic descriptor. The inherent granularity of the global descriptor of this nature arises as a manifestation of the diversity of the locally available descriptors. The granular descriptor can be expressed with the aid of any of the formal models of information granules including sets, fuzzy sets, rough sets, probabilistic granules and others. The architectural essence of the granular descriptor, which supports a quantification of the variability among the sources of knowledge, is realized through an optimal allocation of information granularity. Information granularity is treated as an important design asset and its allocation throughout the parameters of the logic descriptors helps quantify the diversity of individual sources of knowledge. Various protocols of allocation of information granularity along with an overall quantification of their effectiveness are discussed along with their numeric characterization.

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

Corrigendum to “A characterization of electrocardiogram signals through optimal allocation of information granularity” [Artif. Intell. Med. 54 (2012) 125–134

Rural poverty is a hot topic around the world, and the precision of policy making become the key of China's poverty alleviation strategy in the new era. Based on the theory of information granularity in artificial intelligent domain, this study deeply analyzed the massive data from the targeted poverty alleviation big data platform by data mining, and then constructed a Granular poverty differentiation model to identify poverty differentiation characteristics at different levels of granularity in parts of China. We build a knowledge base to support the policy making of targeted poverty alleviation. This result can well support the research of China’s rural poverty.

In the area of optimal asset allocation, the parameters of the model are often assumed to be deterministic. This is not a realistic assumption since most parameters are not known exactly and therefore, have to be estimated. We consider investment opportunities which are modelled as local geometric Brownian motions. The drift terms of the risky assets are assumed to be affine functions of factors. These factors themselves may be stochastic processes. The investor is assumed to have constant relative risk aversion. The optimal asset allocation under partial information is derived by transforming the problem into a full-information problem, where the solution is well known. The analytical result is empirically tested in a real-world application. In our case, we consider the optimal management of a sector rotation fund

Robust supervised rough granular description model with the principle of justifiable granularity

A method of information granule computation is employed to predict the contact resistance for a kind of aircraft electrical relays. The aim is to reduce the dimensional computation complexity and provide practical model for the evaluation process of reliability analysis and fault diagnosis. At first, numerical data sample is split into consecutive sequences by equal window length, and then the sequences are taken to construct information granules, following that, the information granules are combined together to form new time series with reduced dimensionality. Secondly, the fuzzy c-mean clustering method is used as a technique to extract the features of the data points within each information granule. The clustered information granule is defined on the basis of fuzzy linguistic terms to provide parameters for fuzzy inference system, and the fourth order information granular predictive model is constructed in the paper. The predictive model creates the relationship between past four states and future state on the basis of rule base that is produced from fuzzy c-mean clustering. Finally, the predictive model is exploited by introducing experiment data and defining different window length. The trend prediction results are shown and validated with some discussions are presented. The predicative model can reduce the dimension effectively, it is also practical and simple compared with other methods such as neural networks computation and regressive analysis, especially in the case of trend predication application that has highly dimensional measurement data.

Granular modelling of signals: A framework of Granular Computing

The study elaborates on current developments in fuzzy modelling, especially fuzzy rule-based modelling, by positioning them in the general setting of granular computing. This gives rise to granular fuzzy modelling where the models built on a basis of fuzzy models are then conceptually augmented to make them in rapport with experimental data. Two main directions of granular fuzzy modelling dealing with distributed data and collaborative system modelling and transfer knowledge are formulated and the ensuing design strategies are outlined.

In the authors' previous work (1997), an original thermal model of transformers with ONAN cooling was developed. The model takes into account the influence of nonlinear thermal characteristics in transient thermal processes; instead of exponential functions and time constants, the numerical solution of differential equations is used. The model delivers one characteristic temperature in copper (solid insulation) and one characteristic temperature in oil. In this paper, the analysis stresses the definition of the hot-spot temperature of the solid insulation by using easily measurable quantities. Then, the parameters of the model can be precisely determined from inexpensive measurements in a short-circuit heating experiment and the model can deliver the hot-spot temperature. The experimental base of this research are the measurements on a 630-kVA, 3 /spl times/ 10 kV/3 /spl times/ 6 kV ONAN transformer equipped with 112 temperature sensors (102 inside the central positioned 10-kV winding).