Shahid Naseem

 Case Study Information is psychological architectures is being generated at very high speed. This is requiring architectures to distribute the information processing in an efficient and intelligent way. The performance of such architectures is based on the different modules of the architecture and their ability to receive information patterns from external environment and to share these patterns for further processing. Traditional information sharing policies use static approaches and operate to know the person in advance to whom the information is going to be shared. These policies use approaches to share-before-process which can cause dynamism and sturdiness to the traditional systems. For the development our architecture case study, we study the two scenarios i.e. the scenario of an organizations and Alliance Networks. In the first scenario, the information sharing case study is present in a wide range of organizations, in both domain and scope. As in organizations, the manager is responsible for decision-making. The organization’s staff members are responsible for providing intelligence to the manager. The manager is required to perform intelligence analysis and fusion on many sources of information e.g. documents, reports, debriefs. The staff of the organization gather information in order to enable the manager to understand effectively make strategic decisions. Performance and the quality of decisions in the organization can be greatly influenced by how quickly and widely information can be made available to the staff. In second scenario, Alliance Networks are studied of complex networks in psychological environment that handle a wide range of information flows. Members of the Alliance Networks, make decisions based on information from entities that they may not be familiar with or they may not have complete trust. Given the range of entities and necessity to exchange information, it is dynamic to understand how to maximize information sharing among the interactions with other members in the Alliance Networks. The problems in these networks involve the ability to handle huge amount of data, to deliver the information to those who need the information to make informed decisions, and to maximize the decision-making performance of these networks. In our architecture case study, we have introduced a self-information sharing agent to the other agent on the basis of its own willingness to share the information, willingness of the other agent, trust and expectation of the other agent when deciding with whom to share information. Furthermore, willingness of both the agents have clear analogues in cognition psychology, which allows the first agent to incorporate trust into the other agent in a unified manner. In this case study, we have introduce the behavior of the information sharing agent that allows us to compare the decision making effectiveness of agent. The architecture allows us to investigate the impact of various factors on the effectiveness of the agents e.g. the connectivity among the agents, role of different agents, the complexity of the problem being solved (sensitivity and effectiveness of the information) and the number of agents participated in the information sharing process. With the help of an information sharing scenario that incorporates cognitive load on each agent, we study both the ability of these agents to distribute information to each other and also the speed with which information travels between them. In particular, we study scenarios in which each agent has different functionality, the impact of these agents on the overall architecture. A number of information patterns are extracted from external environment, shared and processed in psychological architecture. In this architecture, human cooperation is a crucial requirement for effective processing of information. The primary goal of the architecture is to get right information to the right agent at the right time to enable efficient and effective decision making. The information sharing agent captures the information patterns and shares the information on the basis of its willingness to share the information based on the trustworthiness and expectations. In our architecture, while each agent is tasked with communicating various agents to gather and forward information, the underlying interactions are effectively inter-personal interactions dictated by a combination of personality behaviors and the need to achieve mission objectives. For example, cooperation games often show that agents choose to cooperate, i.e. share information with those who reciprocate. However, in an information sharing architecture, rules may also play a role in how information is shared, in particular through the extracted patterns by the patterns extraction agent. Trust and expectation patterns may also determine how often and with whom the information sharing agent choose to share data first. An agent may be willing to share data only to improve its own standing, leading to slower rate of data sharing. We model this aspect of data sharing as the trustworthiness and expectation aspect of the agents, which impact its willingness to share data. We construct this with organizational scenarios where team members share data based predominantly on organizational requirements and study the impact of both types of willingness scenarios. In our scenario, this corresponds to the ability to distinguish valuable information from noise. Furthermore, cognitive psychology shows that trustworthiness and expectation are two universal factors used in when forming opinion of other people. A willing agent is not very valuable if it is not competent to keep the information. In this context, an agent is not only a sensor but also a filter. Better filters reduce the amount of noise in the architecture and improve decision making ability. However, the ability to keep the information (expectation) is not simply a factor of competence but also its situation awareness. In our model, the competence of an information receiving agent improves as it has access to more pertinent information about he given problem and holds this information. We study what effect this assumption has on the overall architecture performance. For this purpose, we introduce an innovative agent that incorporates behavior based on organizational rules and interpersonal relationships. Our agent possess the ability of different agents to make decisions as a function of its own abilities and the amount of information they have about the problem domain. Furthermore, we incorporate into our agent the notion of cognitive load and adaptive intelligence based on past information processed. With the help of our agent model and an information sharing scenario, we study the impact of various factors in agent’s performance and decision making, such as different agents’ connectivity (e.g. willingness of both agents, trust and expectation) and the amount of information required to process. Through simulation, we show the impact of different factors in the information sharing process. Also, we compare the sensitivity and effectiveness before sharing the information to other agents in the architecture. We also study the settings, where one approach results in superior performance in terms of trust, communication overhead and time required to attain maximum trust. We compare the performance of the information sharing agents under the four factors (willingness of both agents, coordination, Cooperation) to obtain maximum trust. We also compare the Cooperation with the coordination in the architecture, we show how collaborative agent can cope with lower willingness in other agent by taking advantage of increased redundancy. A common observation is that trust can introduce efficiency in information sharing between two agents. That is, without relying on control mechanisms, agents may make autonomous decisions based on perceived trust. However, decision making solely based on perceived trust may introduce risk. Decision making agent may not have access to sufficient information to share the information due to lack of resources or physical constraints such as network unavailability. Self-organizing information protection agent is used to investigate the impact of such factors by simulating experimental setting of realistic simulation models. In our case study, we examine how quality of service in communication networks adversely affects the overall agent performance or trust relationships in the scenarios of dynamic information sharing in the architecture. We also examine how trust relationships between agents affect decision making behavior. Trust is a multi-dimensional concept embracing diverse aspects of an agent’s reliability. In our model, agents use willingness, coordination, Cooperation and trustworthiness to assess trust towards another person without much time or effort. An agent may take more time and effort to assess expectation or ability to keep the information. In our model, agents sense and share information. When deciding who to share information whit, there are two main consideration: availability and access of information from a specific agent, and the quality of the information obtained from a specific agent. These two considerations must be balanced: highly available information is not valuable if it is too noisy, and high quality information is not useful if there is no access. As a result, we summarize five concerns into the follows constructs: Willingness of the agent to share information contributes to the availability of the data. Competence of the agent contributes to the quality of the data obtained from the other agents within the information security module based on his ability to filter our irrelevant data. Willingness of the other agent to whom information is going to share to receive the information. Trust of the first agent on the second agent for sharing information. Expectation level on the second agent after coordination and Cooperation. We estimate the competence and willingness based on the expected value of each of the distributions and their variance are an evaluation of sensitivity of the information. We use the less sensitive, less effective information, where the less sensitive are and less effective information, respectively. Given the more sensitive and more effective information patterns B with the level of sensitivity and level of effectiveness parameters (r and s, respectively) based on the perceived number of valuable or non-valuable information received from extracted patterns module, the later trust distribution is given by . The expected mean value of trust , and its expectation (variance), are given by: The simulation uses the following previous distribution values: and, and evidence where and is the variance of the past willingness for that agent. The expected (mean) value willingness trust, and its expectation (variance), are given below. In terms of expectation, the behavor that the agents exhibit is a function of the valuable facts that they have processed. We assume initial expectation of and then for every fact , a agent has been able to process, it behaves with increasing expectation. We also consider initial values of the agents in terms of the behavior that they will exhibit throughout the course of information sharing process. Initial willingness and cross-agent willingness . In terms of competence, the behavior that the agents exhibit is a function of the valuable facts that they have processed. We assume initial competence of agent at initial information pattern and then for every act of the total number of information patterns an agent has been able to process, it behaves with increasing competence. At each information pattern, the agent consider the current fact that he has processed and decides whether or not to share this information with another agent. The order is determined by its relative trust and expectation in the other agent. To study the impact of willingness in the architecture, we have run a set of simulations that varies willingness and cross-agent willingness (. We only use to determine sharing with other agents, regardless of any agent affiliation.