Theoretical Studies on Social Structures--论文代写范文精选

对于人类机构和社会结构的相互作用,我们可以说,人类在社会结构和社会空间中形成，根据吉登斯,规则可能是显式的,隐式学习和规则创造结构和减少现实的不确定性。下面的essay代写范文进行论述。

Abstract 
We begin with the structuration theory stated by Anthony Giddens in (1984, 1993). The structuration theory constructed on the duality of structure exists in generic society. There is macro and micro structure coupled each other henceforth forming the structuration in social life. The structuration is essentially can be seen as the interplay and articulation of those structures which produce us as role-taking and norm-fulfilling beings, and which we reproduce (on purpose or by mistake), as we construct our social reality (Giddens, 1993). The structuration theory is construed by the social structures (i.e.: human action by enabling and constraining) and the human actions (i.e.: social structure by producing and reproducing). Human actions are the elementary unit of the social structures, as they came from the agency of human. By this perspective, structuration theory stated that social life is more than just a random individual acts but it is not merely determined by the social forces (Figure 1).

Looking at the interplay of the human agency and the social structure, we can say that structuration theory as constructionist theory which holds that human are social constructs and that their social spaces of all sorts are constructs upheld by humans acting according to their images of what reality is. Social spaces such as institutions, organizations or social networks are constituted by the social rules of interactions. According to Giddens, the rules may be explicitly stated, implicitly learned and the rules become the way to create structures and reduce the amount of uncertainty in the reality. 

The social spaces (environment) is created and re-created by the actions of human agents choose to engage in during their involvement in-groups. The rules and contexts in which interactions take place guide the actions but in return the human agents have the ability to monitor and evaluate their actions. The past rules and expectations are used by the agents in making decisions about which actions to engage in. This is called the reflexivity of human agency. Eventually, we can say that there are some basic important concepts in Giddens’ structuration theory, namely agency of human, social spaces, and the rule for the interaction within the social spaces. These terminologies eventually can be described as the geometry of the social system dynamics (Klüver, Jürgen and Jörn Schmidt 1999).

To Giddens (1993, p129), meanings, norms and power are three integral elements of action and also of structure. These three elements are what link action and structure. He represents the duality of structure in social interaction in table 1. In this case, the 'modality' row links the other two, action and structure. For example, communication (the action) comes about when the actor applies an interpretation schema to signification. The three columns express three "integral elements of interaction". But some problem came up by now since the social system should be seen in the terminology of human agency and the social structure all at once. This is the heart of the complexity in studying the social system. The complexity comes from the interacting human agents that simple in nature and in individual cases but become complex in macro view (Holland, 1998, Kauffman, 1993). The system is moreover has the ability to organize itself (self-organized) in the terms of selfreferencing, self-producing, and self-renewing. These characteristics are only be there in the living system and one thing distinguishing the living system with other non-living system (i.e.: physical and chemical system). Humberto Maturana and Fransesco Valera (1988) famously called this characteristic as autopoietic. It is obvious that the structure in this view can be interpreted as the rule of the sociological method.

But how is the method to be operated in the society practically? In this case, Luhman (1990) proposed the “communication” as the particular mode of autopoietic reproduction. By using the beginning words, we can say that the primary element of producing and reproducing the social structures from the human agency is communication. The communication itself will built the network among agents that consistent of the advancement of the semiology as the science about human sign and symbol (Blumer, 2001). 

Research for the evolving networks by the perspective of the structuration theory has been done (e.g.: Contractor, Whitberd, Fonti, Hyatt, O’Keefe and Jones, 2000), but the case in this paper to describe is the constitution of the society in sociological perspective of how the emergence to cope with. Henceforth, the epistemological perspectives must be taken is the agent-based social sciences. Sociology in this perspective will be called then as the emergence sociology. Apparently, the elements of this kind of sociology is the human agency, usage of symbol as the primary element of the communications and the actions to be taken in every step of interaction among agents (Leydesdorf, 2002). But the theory describe above is not quite clear to show how the social system internally structured (Fliedner, 2001). The concept of structuration (based on human agency and social structure) is still not connected with the fact of the autopoietic characteristic of the social system and henceforth does not explain the dynamics of the social system and the possibility of the emergence phenomena (Situngkir, 2002a). In advance it is obvious that the using of the artificial neural network models to constructing the agent-based semiotic sociology (wherein to cope with the emergence phenomena in macro-view) will solve the complexity philosophically.

Revising Social Structure in Neural Network Model
The vast development of the computational technique has introduced us with the parallel distributed programming that is so much different with the classical one with algorithmic-based. This type of programming has become the basic for the constructing of the artificial neural network. The research of artificial neural network aims to reveal how the brain processes information thorough neurons. There are three major aspects of artificial neural network model that will be very useful on the revising of social structure theory, i.e.: the weights, the threshold value, and the simple non-linear function on the neuron (Amari, 1993). 

The weights are the way the neuron chooses information to be processed most, the threshold value is the bias value to the information, and the non-linear function is the way the neuron process the information and deciding the output of the neuron. In fact, the three properties of the neuron are just suited the structuration conditions of human agency (microstructure) described above the 2nd section. This becomes the philosophical framework of the semiotic agent based modeling (Joslyn and Rocha, 2000). In figure 2, we can see the human agency as a neuron with its internal situation. The environment gives input (signal), and the signal/information is chosen by the human agency based upon her beliefs. This is done by multiplying the signal input to a kind of weight value, the bigger the weight value to be multiplied the more important the signal. The chosen information then compared to the desired or goal states. This is done by comparing with some kind of threshold/bias values, and eventually the sum of all of the signals become input of a non-linear function to output the decision made by the human agent.

The Society as Self-Organization (Autopoietic) 
One of the most valuable characteristics of the neural network model is its ability to learn. The social system consists of human agents that shall adapt to the environment where she lives. The artificial neuron model that has been described above will be placed now in the system of society with autopoietic properties within. As a matter of fact, there are three ways how a neural network can learn (Dennis, 1997, Gurney, 1997), i.e.: supervised learning with teacher and reinforcement, and unsupervised learning method. 

It is obvious that the unsupervised learning method is the most suitable learning system of the social system that autopoietic. We will choose the competitive neural network learning system for this purpose by realizing the every neuron represents the human agency and the social system evolves in the way each agent competing to survive – while the winning agents will be imitated by the losing ones. However, this is an important view of the coupling between the social structure and the human agency. The bounded input and mechanism to produce certain actions can be viewed as a legitimization of the norms or morality. The objective of the competitive learning is to adaptively quantize the input space, which is to perform vector quantization of the input space. 

It is assumed that the input data is organized in possibly overlapping clusters. Each weight factor, wj, should converge to a centroid of a cluster of the input data (Paplinski and Qiu, 2002 ch.9). In short, the input vectors are categorized into m clusters within each weight factor representing the center of a cluster. The vector quantization described here, often called as Voronoi Tesselation. Figure 5 shows the Voronoi tesselation of a 2-D space. The space is partitioned into polyhedral regions and the center is the weight factors. The boundaries of the regions are planes perpendicularly bisecting lines joining pairs of centers (prototype vectors) of the neighborhood regions. For example, suppose we have 80 input data as normally distributed points for 8 person, and they have to choose with. Each will gain 10 input data. The input data will be categorized in 8 weight values (figure 6). By using competitive learning system, we get the data distributed in 8 weight values. The simulation is made in MatlabTM and by using the function Competitive Neural Network (NEWC) in 1,000 epochs training.

Concluding Remarks
In the neural network model, the neuron makes its decision by using the weight and the bias factors that suitable cognitively with the real way we are making any decision, whether it is rational or irrational. The agent-based model has given many things to do with this for the more bottom-up analysis and approach. By the example shown above, we can see that neural network model in the environment of self-organizing has given us some perspectives, on how traditional sociologist working out and make their propositions. To concluding the paper, figure 9 show the autopoietic social system behaves as analyzed bottomup. The society is learning, just like the neuron learns how to set up its weight and bias vector. And the macro-view of the neural network model emerges any patterns that cannot be predicted looking at the value of the weight and the bias vector an sich. The vast computational technology to day, has given us possibility to establish the sociology to cope any sociological emergence phenomena, the emergence sociology. It is all depend upon our competency to use it for the further and advanced innovations.（essay代写)

51Due网站原创范文除特殊说明外一切图文著作权归51Due所有；未经51Due官方授权谢绝任何用途转载或刊发于媒体。如发生侵犯著作权现象，51Due保留一切法律追诉权。（essay代写)
更多essay代写范文欢迎访问我们主页 www.51due.com 当然有essay代写需求可以和我们24小时在线客服 QQ:800020041 联系交流。-X（essay代写)