Phase transition in SONFIS&SORST

📝 Abstract

In this study, we introduce general frame of MAny Connected Intelligent Particles Systems (MACIPS). Connections and interconnections between particles get a complex behavior of such merely simple system (system in system).Contribution of natural computing, under information granulation theory, are the main topics of this spacious skeleton. Upon this clue, we organize two algorithms involved a few prominent intelligent computing and approximate reasoning methods: self organizing feature map (SOM), Neuro- Fuzzy Inference System and Rough Set Theory (RST). Over this, we show how our algorithms can be taken as a linkage of government-society interaction, where government catches various fashions of behavior: solid (absolute) or flexible. So, transition of such society, by changing of connectivity parameters (noise) from order to disorder is inferred. Add to this, one may find an indirect mapping among financial systems and eventual market fluctuations with MACIPS. Keywords: phase transition, SONFIS, SORST, many connected intelligent particles system, society-government interaction

💡 Analysis

In this study, we introduce general frame of MAny Connected Intelligent Particles Systems (MACIPS). Connections and interconnections between particles get a complex behavior of such merely simple system (system in system).Contribution of natural computing, under information granulation theory, are the main topics of this spacious skeleton. Upon this clue, we organize two algorithms involved a few prominent intelligent computing and approximate reasoning methods: self organizing feature map (SOM), Neuro- Fuzzy Inference System and Rough Set Theory (RST). Over this, we show how our algorithms can be taken as a linkage of government-society interaction, where government catches various fashions of behavior: solid (absolute) or flexible. So, transition of such society, by changing of connectivity parameters (noise) from order to disorder is inferred. Add to this, one may find an indirect mapping among financial systems and eventual market fluctuations with MACIPS. Keywords: phase transition, SONFIS, SORST, many connected intelligent particles system, society-government interaction

📄 Content

Phase transition in SONFIS&SORST Hamed Owladeghaffari

Department of Mining&Metallurgical Engineering Amirkabir University of Technology Tehran,Iran

h.o.ghaffari@gmail.com
Abstract. In this study, we introduce general frame of MAny Connected Intelligent Particles Systems (MACIPS). Connections and interconnections between particles get a complex behavior of such merely simple system (system in system).Contribution of natural computing, under information granulation theory, are the main topics of this spacious skeleton. Upon this clue, we organize two algorithms involved a few prominent intelligent computing and approximate reasoning methods: self organizing feature map (SOM), Neuro- Fuzzy Inference System and Rough Set Theory (RST). Over this, we show how our algorithms can be taken as a linkage of government-society interaction, where government catches various fashions of behavior: “solid (absolute) or flexible”. So, transition of such society, by changing of connectivity parameters (noise) from order to disorder is inferred. Add to this, one may find an indirect mapping among finical systems and eventual market fluctuations with MACIPS. Keywords: phase transition, SONFIS, SORST, many connected intelligent particles system, society-government interaction 1 Introduction Complex systems are often coincided with uncertainty and order-disorder transitions. Apart of uncertainty, fluctuations forces due to competition of between constructive particles of system drive the system towards order and disorder. There are numerous examples which their behaviors show such anomalies in their evolution, i.e., physical systems, biological, financial systems [1]. In other view, in monitoring of most complex systems, there are some generic challenges for example sparse essence, conflicts in different levels, inaccuracy and limitation of measurements ,which in beyond of inherent feature of such interacted systems are real obstacle in their analysis and predicating of behaviors. There are many methods to analyzing of systems include many particles that are acting on each other, for example statistical methods [2], Vicsek model [3]. Other solution is finding out of “main nominations of each distinct behavior which may has overlapping, in part, to others”. This advance is to bate of some mentioned difficulties that can be concluded in the “information granules” proposed by Zadeh [4]. In fact, more complex systems in their natural shape can be described in the sense of networks, which are made of connections among the units. These units are several facets of information granules as well as clusters, groups, communities, modules [5]. Let us consider a more real feature: dynamic natural particles in their inherent properties have (had have-will have) several appearances of “natural” attributes as in individually or in group forms. On the other hand, in society, interacting of main such characteristics (or may extra- natural forces: metaphysic) in facing of predictable or unpredictable events, determines destination of the supposed society.
Based upon the above, hierarchical nature of complex systems [6], developed (developing) several branches of natural computing (and related limbs) [7], collaborations, conflicts, emotions and other features of real complex systems, we propose a general framework of the known computing methods in the connected (or complex hybrid) shape, so that the aim is to inferring of the substantial behaviors of intricate and entangled large societies. Obviously, connections between units of computing cores (intelligent particles) can introduce part (or may full) of the comportments (demeanors-deportments…). Complexity of this system, called MAny Connected Intelligent Particles Systems (MACIPS), add to reactions of particles against information flow, and can open new horizons in studying of this big query: is there a unified theory for the ways in which elements of a system(or aggregation of systems) organize themselves to produce a behavior?[8]. In this study, we select a very little limited part of MACIPS (fig.1.), as well as hybrid intelligent systems, and investigate several levels of responses in facing of real information. We show how relatively such our simple methods that can produce (mimic) complicated behavior such government-nation interactions. Mutual relations between algorithms layers identify order-disorder transferring of such systems. So, we found our proposed methods have good ability in mimicking of government-nation interactions while government and society can take the different states of responses. Developing of such intelligent hierarchical networks, investigations of their performances on the noisy information and exploration of possible relate between phase transition steps of the MACIPS and flow of information in to such systems are new interesting fields, as well in various fields of science and econo

This content is AI-processed based on ArXiv data.