ADMIT

Agent-Oriented Distributed Data Mining using Computational Statistics

 Partecipanti

Mappa

 Word cloud

Esplora la "nuvola delle parole (Word Cloud) per avere un'idea di massima del progetto.

 Obiettivo del progetto (Objective)

'Today's systems for managing critical infrastructure such as traffic, energy, or industry automation systems are highly complex, distributed, and increasingly decentralized. Multi-agent systems (MAS) provide an intuitive metaphor and configurable, robust and scalable methods for problem-solving and control in distributed, decentrally organized system. The purpose of Distributed Data Mining (DDM) is to provide algorithmic solutions for data analysis in a distributed manner to detect hidden patterns in data and extract knowledge necessary for decentralized decision making. A new promising area of research studies possibilities for coupling MAS and DDM by exploiting DDM methods for improving agents’ intelligence and MAS systems performance. In the ADMIT project we focus on methods for distributed estimation of parameters for the individual agents, agent communities, and application-level information models. Our approach is based on Computational statistics (CST), which includes a set of methods for approximate solution of statistical problems without complex statistical procedures. The goal of the ADMIT project is to develop an agent-oriented DDM framework, which includes a set of computationally effective, robust and easy to apply methods for models parameter estimation and allows easy incorporation into MAS applications to analyze models at different levels of MAS. The scientific research objectives of ADMIT are: 1. To develop a conceptual architecture of agent-oriented DDM framework as well as a methodology of its usage in the multiagent programming frameworks; 2. To develop a set of computationally effective and reliable to bad data quality CST-based DDM methods, for efficient estimation of the models parameters on the basis of distributed data as well as estimate the methods performance; 3. To access the impact of incorporation of the DDM framework to MAS-based applications (with the main focus on traffic and logistics domains).'

Introduzione (Teaser)

In recent years, several approaches have been adopted to reveal valuable knowledge hidden in the huge amount of data produced everyday in many fields of science and information technology applications. A new solution proposed by EU-funded researchers makes use of intelligent software agents.

Descrizione progetto (Article)

Among the steps to follow in the process of extracting unknown knowledge out of large databases, data mining is a central element. Techniques to automate the extraction of repeating patterns were initially developed for centralised data. As industry and science increasing rely on geographically dispersed computing resources, methods for distributed data mining started to emerge.

In fact, numerous solutions are available using techniques such as distributed clustering, classification and regression. However, only a few of them rely on intelligent agents that can control a growing number of data mining tasks. The EU-funded project 'Agent-oriented distributed data mining using computational statistics' (http://www.in.tu-clausthal.de/personen/aktuelle/dr-jelena-fiosina/admit/ (ADMIT)) explored the added value of concepts borrowed from agent technology.

In multi-agent systems, the individual and collective behaviours of agents depend on the observed data. ADMIT researchers considered decentralised data processing techniques, including regression forecasting and change-point analysis to determine if and when a change in a data set has occurred. These data coordination models were applied for decision making and achieved similar performance as with a central authority.

?he synergy between such communities of agents and cloud computing offered additional perspectives for new technologies. ADMIT researchers examined decentralised data clustering that is an important data pre-processing step in cloud data repositories. By grouping similar data together, it was possible to construct more accurate data representatives for application such as optimal route selection and speed adaptation in traffic.

ADMIT's many results were presented at 11 international conferences. Sixteen papers were also published in conference proceedings and peer-reviewed scientific journals. More importantly, the proposed methods have been integrated into intelligent transportation systems for traffic management and environment monitoring, and validated using real-world traffic data from the German city of Hannover.