Cooperative Multi-Agent Joint Action Learning Algorithm (CMJAL) for Decision Making in Retail Shop Application

This article gives a novel approach to cooperative decision-making algorithms by Joint Action learning for the retail shop application. Accordingly, this approach presents three retailer stores in the retail marketplace. Retailers can help to each other and can obtain profit from cooperation knowledge through learning their own strategies that just stand for their aims and benefit. The vendors are the knowledgeable agents to employ cooperative learning to train in the circumstances. Assuming a significant hypothesis on the vendor's stock policy, restock period, and arrival process of the consumers, the approach was formed as a Markov model. The proposed algorithms learn dynamic consumer performance. Moreover, the article illustrates the results of cooperative reinforcement learning algorithms by joint action learning of three shop agents for the period of one-year sale duration. Two approaches have been compared in the article, i.e. multi-agent Q Learning and joint action learning.

Adaptive Congestion Controlled Multipath Routing in VANET

This article describes how in the VANET environment, routes are broken owing to node mobility. Moreover, the usage of wireless links for data communication leads to inherent unreliability and are error prone. Single path routing uses a prediction mechanism to compute a reliable path considering vehicle velocity and vehicle direction. Nevertheless, this methodology does not deal with major real-world traffic conditions. Hence, to address the aforementioned problems and to enhance reliability and fault tolerance, multipath routing protocols are employed. However existing multipath routing protocols even though compute multipath, only one path will be engaged in actual communication at any given time. Hence this work proposes Adaptive Congestion Controlled Multipath Routing in a VANET. The proposed work computes multiple paths from source to destination using cubic Bezier curves and more importantly, employs all/more than one path during the communication. The paths thus computed are adaptive in nature dependent upon the direction of mobility of source and destination vehicles.

Artificial Minds with Consciousness and Common sense Aspects

The research work presented in this article investigates and explains the conceptual mechanisms of consciousness and common-sense thinking of animates. These mechanisms are computationally simulated on artificial agents as strategic rules to analyze and compare the performance of agents in critical and dynamic environments. Awareness and attention to specific parameters that affect the performance of agents specify the consciousness level in agents. Common sense is a set of beliefs that are accepted to be true among a group of agents that are engaged in a common purpose, with or without self-experience. The common sense agents are a kind of conscious agents that are given with few common sense assumptions. The so-created environment has attackers with dependency on agents in the survival-food chain. These attackers create a threat mental state in agents that can affect their conscious and common sense behaviors. The agents are built with a multi-layer cognitive architecture COCOCA (Consciousness and Common sense Cognitive Architecture) with five columns and six layers of cognitive processing of each precept of an agent. The conscious agents self-learn strategies for threat management and energy level maintenance. Experimentation conducted in this research work demonstrates animate-level intelligence in their problem-solving capabilities, decision making and reasoning in critical situations.

Virtual Enterprise as a Multi Agent System

This article describes changing customer demands require that enterprises mobilize their resources to quickly develop a suitable product. This is achievable if competing enterprises collaborate to deliver the product. Each of them brings their expertise into the collaboration. This collaboration where each enterprise brings in its core competency is referred to as a virtual enterprise (VE). A construction project is implemented by a team of professionals and an alliance of companies that is formed by consultants who evaluate contractors for specific project tasks. Partners can be represented as multiple agents. Prior evidence of multi-agent system (MAS) model that facilitates formation of VEs is lacking. VE MAS ontology has been designed and used in agent interactions. The model can be used in evaluation and selection process of partners. Delegation of the process to the model, gives partners time to implement the tasks. Partner evaluation and selection problem for building construction projects is solvable if pragmatic scientific approaches are employed with appropriate mathematical models. This article proposed a VE model for evaluating and selecting right partners for building construction projects. The model was used to demonstrate the choice of the most preferred partner. Researchers have not evaluated this model but propose that once in place, it can evaluated against manual selection of potential partners using similar parameters by examining the closeness of the output.

Multi-Agents Machine Learning (MML) System for Plagiarism Detection

Day after day the cases of plagiarism increase and become a crucial problem in the modern world caused by the quantity of textual information available in the web. Data mining becomes the foundation for many different domains as one of its chores is the text categorization, which can be used in order to resolve the impediment of automatic plagiarism detection. This article is devoted to a new approach for combating plagiarism named MML (Multi-agents Machine learning system) and is composed of three modules: data preparation and digitalization, using n-gram character or bag of words as methods for the text representation; TF*IDF as weighting to calculate the importance of each term in the corpus in order to transform each document to a vector; and learning and voting phase using three supervised learning algorithms (decision tree c4.5, naïve Bayes and support vector machine).

The Meaningfulness of Statistical Significance Tests in the Analysis of Simulation Results

This article discusses the question of whether significance tests on simulation results are meaningful at all. It is also argued that it is the effect size much more than the existence of the effect is what matters. It is the description of the distribution function of the stochastic process incorporated in the simulation model which is important. This is particularly when this distribution is far from normal, which is particularly often the case when the simulation model is nonlinear. To this end, this article uses three different agent-based models to demonstrate that the effects of input parameters on output metrics can often be made “statistically significant” on any desired level by increasing the number of runs, even for negligible effect sizes. The examples are also used to give hints as to how many runs are necessary to estimate effect sizes and how the input parameters determine output metrics.

An Agent-Based Model of Insurance and Protection Decisions on IT Systems

This paper discusses the key role of incentives in information systems security. Vulnerabilities can be reduced, and even removed, if individual motivations are taken into account in the process of protection and insurance design. The article first discusses the importance of externalities, free-riding behavior, uncertainty and the incentives mismatch between individuals and organizations involved in information systems security. Previous works perform this study using a game theoretical approach but the paper shows that an agent-based model is capable of including the heterogeneity and interrelations among individuals, not focusing on the reached equilibrium but on the dynamics prior to its emergence.

A Novel Approach to Change Management in Requirements Engineering Context

Requirements change management (RCM) is actually an inevitable task that might be considered in system development's life cycle, since user requirements are continuously evolving (some are added, others are modified or deleted). A large majority of studies have examined the issue of change, while most of them focused on the design and source code, requirements were often forgotten, even though, the cost of fixing the defect and introduced error due to the requirements is less higher compared to the cost of error in design and implementation. For this purpose, this work focuses on change issues in the requirements engineering (RE) context, which contains the complete initial specification. Properties such as adaptability, perception, and cooperation of the multi-agent system (MAS) allow managing changing requirements in a controlled manner. The main objective of this work is to develop an agent-oriented approach which will be effective in the requirements management to be adapted to changes in their environments.

(HR)^2

Finding the right candidate for a job has always been a hard task that Human Resources (HR) managers of a company face regularly. In this paper, the authors propose that the field of multi-agents can play a significant role in a) elaborating the job description b) getting an applicant to submit competencies relevant to the job c) shortlisting applicants and d) identifying the right hire. They propose the model of (HR)^2, an automated agent for Helping HR with Recruitment that could perform the following key steps: (a) Generate Specific Position Contract (SPC) from a Master Position Contract (MPC) using Infer1 procedure (b) Use the SPC to provide a graded and iterative feedback to applicant using Infer2 procedure. They situate (HR)^2 in the context of LinkedIn. To enable better inference, they propose to modify the information being collected by LinkedIn, using the ontology provided by the free online database O*NET. The (HR)^2 agent will be able to help the employer rank order the SPCs and identify areas for assessment, potentially easing the interview process and leading to high quality hires.

Assessing the Effect of Aspect Refactoring on Multi-Agent Applications

Multi Agent Systems (MAS) are increasingly gaining importance as a powerful paradigm to designing and implementing distributed applications. However, existing multi-agent applications are developed without considering the separation of non-functional concerns from the functional ones. This makes the implementation, comprehension and maintenance of multi-agent applications hard tasks. Aspect-Oriented Refactoring (AOR) is a promising technique for improving modularity and reducing complexity of existing object oriented software systems by encapsulating crosscutting concerns. The authors present, in this paper, a new dynamic approach for investigating empirically the effect of AOR on MAS applications. They focus, particularly, on the effect of AOR on agent behavior in terms of communication. The proposed approach is supported by a multi-agent profiling tool working on AgentFactory platform.