Vladimir Gorodetsky

P2P Agent Platform: Implementation and Testing

Peer-to-Peer (P2P) computing, a novel paradigm for distributed information technology, is currently receiving ever increasing interest from both academia and industry. Recent efforts undertaken to integrate multi-agent and P2P architectures are one of such very promising new opportunities. Indeed, multi-agent system (MAS) may be thought of as a multitude of autonomous entities, and, therefore, structuring the agents as a P2P network of nodes may result in an architecture providing a new dimension for design of open MAS composed of a highly transient population of agents. This paper presents an implementation of a P2P Agent Platform providing transparent interaction for distributed P2P agents. The developed P2P Agent Platform implements the basic mandatory components assumed by the functional architecture proposed by the FIPA Nomadic Agents Working Group. This implementation is supported with a search mechanism, a function of an underlying P2P infrastructure. The platform verification is done via prototyping a P2P ground object detection MAS in which the agents situated on top of the distributed instances of the P2P Agent platform provide classification services.

Autonomous Intelligent Systems: Multi-Agents and Data Mining

Invited Talks.- Peer-to-Peer Data Mining, Privacy Issues, and Games.- Ontos Solutions for Semantic Web: Text Mining, Navigation and Analytics.- Robust Agent Communities.- WI Based Multi-aspect Data Analysis in a Brain Informatics Portal.- Agent and Data Mining.- Agent-Mining Interaction: An Emerging Area.- Evaluating Knowledge Intensive Multi-agent Systems.- Towards an Ant System for Autonomous Agents.- Semantic Modelling in Agent-Based Software Development.- Combination Methodologies of Multi-agent Hyper Surface Classifiers: Design and Implementation Issues.- Security in a Mobile Agent Based DDM Infrastructure.- Automatic Extraction of Business Rules to Improve Quality in Planning and Consolidation in Transport Logistics Based on Multi-agent Clustering.- Intelligent Agents for Real Time Data Mining in Telecommunications Networks.- Architecture of Typical Sensor Agent for Learning and Classification Network.- Self-organizing Multi-agent Systems for Data Mining.- Role-Based Decision Mining for Multiagent Emergency Response Management.- Agent Competition and Data Mining.- Virtual Markets: Q-Learning Sellers with Simple State Representation.- Fusion of Dependence Networks in Multi-agent Systems - Application to Support Net-Enabled Littoral Surveillance.- Multi-agent Framework for Simulation of Adaptive Cooperative Defense Against Internet Attacks.- On Competing Agents Consistent with Expert Knowledge.- On-Line Agent Teamwork Training Using Immunological Network Model.- Text Mining, Semantic Web, and Agents.- Combination of Rough Sets and Genetic Algorithms for Text Classification.- Multi-agent Meta-search Engine Based on Domain Ontology.- Efficient Search Technique for Agent-Based P2P Information Retrieval.- Classification of Web Documents Using Concept Extraction from Ontologies.- Emotional Cognitive Agents with Adaptive Ontologies.- Viral Knowledge Acquisition Through Social Networks.- Chinese Weblog Pages Classification Based on Folksonomy and Support Vector Machines.

Agent-Based Distributed Data Mining: A Survey

Distributed data mining is originated from the need of mining over decentralised data sources. Data mining techniques involving in such complex environment must encounter great dynamics due to changes in the system can affect the overall performance of the system. Agent computing whose aim is to deal with complex systems has revealed opportunities to improve distributed data mining systems in a number of ways. This paper surveys the integration of multi-agent system and distributed data mining, also known as agent-based distributed data mining, in terms of significance, system overview, existing systems, and research trends.

Guest Editors' Introduction: Agents and Data Mining

On top of two active research streams, agents and data mining, a most recent and exciting trend is their interaction and integration. Agent mining has emerged as a very promising field due to its unique contributions to complementary and innovative methodologies, techniques, and applications for complex problem-solving. This editorial summarizes the structure of this special issue.

Asynchronous alert correlation in multi-agent intrusion detection systems

This paper presents conceptual model, architecture and software prototype of a multi-agent intrusion detection system (IDS) operating on the basis of heterogeneous alert correlation. The latter term denotes IDS provided with a structure of anomaly detection–like classifiers designed for detection of intrusions in cooperative mode. An idea is to use a structure of classifiers operating on the basis of various data sources and trained for detection of attacks of particular classes. Alerts in regard to particular attack classes produced by multiple classifiers are correlated at the upper layer. The top-layer classifier solves intrusion detection task: it combines decisions of specialized alert correlation classifiers of the lower layer and produces combined decision in order to more reliably detect an attack class. IDS software prototype operating on the basis of input traffic is implemented as multi-agent system trained to detect attacks of classes DoS, Probe and U2R. The paper describes structure of such multi-layered intrusion detection, outlines preprocessing procedures and ‘data sources, specifies the IDS multi-agent architecture and presents briefly the experimental results received on the basis of DARPA-98 data, which generally confirm the feasibility of the approach and its certain advantages.

On-Line Update of Situation Assessment Based on Asynchronous Data Streams

The subject of the paper is multi-agent architecture of and algorithmic basis for on-line situation assessment update based on asynchronous streams of input data received from multiple sources and having finite “life time”. A case study from computer network security area that is anomaly detection is used for demonstration.

Big Data: Opportunities, Challenges and Solutions

The problems related to the phenomenon of Big Data are currently among the top 10 hottest topics of information and communication technology. Big Data phenomenon refers to the data explosion observed today. At present, the term is widely used in different communities of many application domains, including researchers and practitioners. Big Data analysis can provide for many new opportunities in many respects motivating and stimulating industrial and commercial take-up of novel emerging technologies. The in-depth analysis of Big Data processing and analytics publications shows that the most of them write about “new opportunities” and “new challenges”. However, very few papers present the solutions for predictive analytics that go beyond the limits of OLAP-like processing models and technologies. The goal of this paper is to outline in more detail not only the nature of opportunities and particular challenges but also some original solutions to attack them.

Agents and Distributed Data Mining in Smart Space: Challenges and Perspectives

Smart space is a distributed ambient environment with existing, inside it, dynamic set of inhabitants (living and nonliving) solving various own and common tasks. The mission of smart space is to provide, for its inhabitants, with context–dependent information, communication, services, reminders and personalized recommendations in a user–friendly mode where and when needed in ubiquitous and unobtrusive style. The smart space R&D uses large diversity of models, frameworks, and technologies and their integration is the first challenging smart space problem. Another challenge is caused by the necessity to process huge volumes of heterogeneous information perceived by distributed sensors in adaptive, self–organizing, learnable, and efficient style. The paper analyses these challenges and emphasizes an important role of the technology integrating agent and data mining to overcome both these challenges.

Ontology-Based Context-Dependent Personalization Technology

Personalization, a topmost concern of modern recommendation systems (RS), is intended to predict individual motivation of a customer for this or that choice. It depends on many factors forming explicit and implicit decision context. The paper proposes RS personalization technology that focuses on ontology–based extraction of semantically interpretable context of each particular customer’s decisions from his/her historical data sample with the subsequent machine learning–based extraction of customer–centered feature set and personal cause–consequence decision rules. The technology is fully implemented by Practical Reasoning, Inc. and validated via several case studies.

Multi-agent and Data Mining Technologies for Situation Assessment in Security-related Applications

The paper considers one of the topmost security related problems that is situation assessment. Specific classification and data mining issues associated with this task and methods of their solution are the subjects of the paper. In particular, the paper discusses situation assessment data model specifying situation, approach to learning of situation assessment, generic architecture of multi-agent situation assessment systems and software engineering issues. Detection of abnormal use of computer network is a case study used for demonstration of the main research results.