ADMIT

ADMIT as a FP7 funded Project

This is FP7 Marie-Curie Actions Intra-European Fellowships (IEF) Project.

ADMIT Project is executed since 1-st June 2011 - 18th July 2014 by experienced researcher Dr. Jelena Fiosina at Clausthal University of Technology, at the Department of Informatics, at the Chair for Business Information Technology in the Mobile and Enterprise Computing research group and supervised by the director of the Department of Informatics and the head of the Mobile and Enterprise Computing research group Prof. Jörg P. Müller.

Nach oben

Abstract

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 groups, and system-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

Objective 1: To develop a conceptual architecture of agent-oriented DDM framework as well as a methodology of its usage in multi-agent programming frameworks;

Objective 2: To develop a set of computationally effective and reliable to bad data quality CST-based DDM methods, for efficient estimation of model parameters on the basis of distributed data as well as estimate the methods performance;

Objective 3: To assess the impact of incorporation of the DDM framework to MAS-based applications (with the main focus on traffic and logistics domains).

    Summary of Progress and Details for each Task

    Objective1 (Tasks 1-2): To develop a conceptual architecture of an agent-oriented (AO) distributed data mining (DDM) framework and a methodology of usage in the MAS frameworks, the necessary data flows and use cases were described and analysed. Considering traffic routing problems, several scenarios were described involving tactical and strategic planning of traffic participants’ behaviour [14, 15]. The necessary data flows for the comparison of routes [1, 8, 12], change-point analysis of traffic state [5, 11], and travel time forecasting [6, 9, 10] and clustering [2, 3, 4] were investigated. The necessary AO models for DDM layer were developed that insure the intelligent behaviour of agents and their collaboration. The conceptual decentralized, distributed and centralized architectures of the AO DDM framework are described in [1-10, 14, 15]. Finally, such Future Internet capability as cloud-computing architecture was considered for decentralized traffic management system, including the concrete scenarios and distributed data flow processing and mining methods [1, 4, 7,].

     

    Objective 2 (Tasks 3-4): A set of computational statistics (CST) based methods for the efficient estimation of model parameters on the basis of distributed data sources was developed, which are computationally effective and reliable to bad data quality. The performance of the methods was estimated. In [14, 15] CST-based methods were applied to decision making strategies. In [1, 8, 12] a resampling based approach to the comparison of routes in a graph was suggested and its efficiency was estimated. In [5] a change-point problem that supports agents in detecting changes in their environment was described and CST-based solution methods were developed. The performance of these methods was analyzed in [11]. CST-based methods were developed to support agent cooperation for forecasting with regression model [4, 6, 9, 10] and for clustering [2,3] of travel times by intelligent agents;

     

    Objective 3 (Tasks 5-6): The impact of incorporating the DDM framework into MAS-based applications in traffic and logistics domains was assessed. All the scenarios [1-10, 14, 15] were taken from the domains mentioned. All methods were validated using experiments using data from real traffic networks. These initial data were obtained from the on-going project PLANETS of NTH, where our research group at TU Clausthal participates. This collaboration with other MEC Group colleagues implied the co-authored publication of Dr. Fiosina.

     

    The results of these objectives were presented by Dr. Fiosina and co-authors at 11 international conferences. 16 project-related papers were published during the reported period (See the next section).

     

    The most significant scientific results

    The scientific results were presented by Dr. Fiosina and co-authors at 11 international conferences. 16 project-related papers in peer-reviewed journals and conference proceedings were published during the reported period. To show evidence for the quality and efficiency of the proposed methods the experimental project result were integrated into traffic domain application use-cases and validated using real-world traffic data from the southern part of Hannover. 

    Result 1: A traffic routing problem with decentralized decision making of vehicle agents in urban traffic system was investigated, where the planning process for a vehicle agent is separated into two stages: strategic planning for selection of the optimal route and tactical planning for passing the current street in the optimal manner. A multi-agent architecture for this problem was developed [14, 15]; data flows and scenarios were analyzed. Necessary CST-based DDM algorithms for comparing two routes in a stochastic graph [8, 12] and the shortest path search were developed [14, 15], which are carried out at strategic planning stage; the efficiency of the algorithms was evaluated.

    Result 2: Change Point Analysis for data processing and data mining of intelligent agents in city traffic was investigated [5]. The necessary agent-oriented architectures, scenarios and data flows were described [11]. Two CST-based resampling tests for change point detection were suggested, which were implemented at the DDM layer of agent logics. 

    Result 3: A problem of decentralised travel time forecasting was considered [4, 6, 9, 10]. A MAS architecture with autonomous agents was implemented for this purpose. A decentralised linear [4, 6, 10] and kernel-based [4, 6, 9] multivariate regression models were developed to forecast the travelling time. The iterative least square estimation method was used for regression parameter estimation, which is suitable for streaming data processing. The resampling-based consensus method was suggested for coordinated adjustment of estimates between neighboring agents. The prediction technique in tutorial style was described as an invited chapter ‘Cooperative Regression-based Forecasting in Distributed Traffic Networks’ of the CRC Press, Taylor and Francis Group book ‘Distributed Network Intelligence, Security and Applications’.

    Result 4: A problem of kernel-density-based clustering was considered [2, 3, 4]. The CST kernel-based methods and a MAS architecture were implemented. Two decentralised clustering models were proposed. The first model [2] is non-parametric based on kernel density functions and uses the cooperation stage between the local model parts. The cooperation assumes the transmission of selected data points to adjust the model parts of different autonomous agents. The second model [3] is semi-parametric based both on kernel-density functions and am approximated density of multivariate normal distributions. The cooperation assumes the transmission of the estimated parameters of a mixture of multivariate normal distributions between the cooperating autonomous agents.

    Result 5: A major result of the project was the development of a strategic personal research agenda with focus on Big Data analysis in conjunction with the use of cloud resources in the domain of traffic information systems. Initial work towards this agenda was performed and several papers published describing challenges and solution approaches in this area. A research grant proposal ANACONDA (Decentralised Big Data Analysis in Complex Networked Applications) was prepared and submitted to Deutsche Forschungsgemeinschaft (DFG) in July 2014. If successful, this grant will supply the research funding for three years and play a crucial role in helping Dr. Fiosina to complete her Habilitation.

     

     

    Project related publications

    Publications published during the reporting period (funded by ADMIT)

    [1] Fiosina, J., Fiosins, M. (2014). Resampling based Modelling of Individual Routing Preferences in a Distributed Traffic Network. Int. Journal of Artificial Intelligence 12(1), pp. 79-103. CESER Publications [Preprint]

    [2] Fiosina, J., Fiosins, M. (2013). Density-Based Clustering in Cloud-Oriented Collaborative Multi-Agent Systems. (HAIS2013), Lecture Notes in Computer Science, Vol. 8073, pp 639-648 [Preprint]

    [3] Fiosina, J., Fiosins, M., Müller, J.P. (2013). Decentralised Cooperative Agent-based Clustering in Intelligent Traffic Clouds. (MATES2013), Lecture Notes in Computer Science Vol. 8076 pp 59-72 [Preprint]

    [4] Fiosina, J., Fiosins, M., Müller, J.P. (2013). Big Data Processing and Mining for Next Generation Intelligent Transportation Systems. Jurnal Teknologi (Sciences & Engineering), 63(3): 23–38, Penerbit UTM Press, Universiti Teknologi Malaysia.[PDF]

    [5] Fiosina, J. and Fiosins, M. (2013). Chapter 1: ‘Cooperative Regression-Based Forecasting in Distributed Traffic Networks’.Qurban A. Memon, ed.,"Distributed Network Intelligence, Security and Applications". CRC Press, Taylor & Francis Group, pp. 3-37. [Preprint]

    [6] Fiosina, J., Fiosins, M. and Müller, J. P. (2013). Mining the Traffic Cloud: Data Analysis and Optimization Strategies for Cloud-Based Cooperative Mobility Management. In Proc. of the Int. Sym. on Management Intelligent Systems (ISMiS2013), Advances in Intelligent Systems and Computing, 220, Springer Verlag, Berlin / Heidelberg, pp. 25-32 , DOI: 10.1007/978-3-319-00569-0_4 [Preprint]

    [7] Fiosina, J. and Fiosins, M. (2013). Selecting the Shortest Itinerary in a Cloud-Based Distributed Mobility Network. In. Proc. of the 10th Int. Conf. on Distributed Computing and Artificial Intelligence (DCAI 2013), Advances in Intelligent Systems and Computing, Vol. 217, Springer Verlag, Berlin / Heidelberg, pp. 103-110, DOI: 10.1007/978-3-319-00551-5_13 [Preprint]

    [8] Fiosins, M., Fiosina, J. and Müller, J. P. (2012). Change Point Analysis for Intelligent Agents in City Traffic. In Cao. L., Bazzan. A., Symeonidis. A., Gorodetsky. V., Weiss. G. and Yu. P. (eds.). Agents and Data Mining Interaction. Lecture Notes in Artificial Intelligence 7103, Springer, pp. 195-210, DOI: 10.1007/978-3-642-27609-5_13, Online ISBN: 978-3-642-27609-5, [Preprint]

    [9] Fiosina, J. and Fiosins, M. (2012). Cooperative Kernel-Based Forecasting in Decentralized Multi-Agent Systems for Urban Traffic Networks. In Proc. of Ubiquitous Data Mining (UDM) Workshop at the 20th European Conf. on Artificial Intelligence, CEUR-WS.org, Vol. 960, pp. 3-7. [Preprint]

    [10] Fiosina, J. (2012). Decentralised Regression Model for Intelligent Forecasting in Multi-agent Traffic Networks. In Proc. of the 9th Int. Conf. on Distributed Computing and Artificial Intelligence (DCAI’12), 28-30 Mar. 2012, Salamanca, Spain, Springer Verlag, S. Omatu et al. (Eds.): Advances in Intelligent and Soft Computing, 151, Springer, pp. 255-263. DOI: 10.1007/978-3-642-28765-7_30, Online ISBN: 978-3-642-24800-9 [Preprint].

    [11] Fiosina, J. and Fiosins, M. (2011). Resampling-based Change Point Estimation. In Proc. of the 10th Int. Sym. on Intelligent Data Analysis (IDA'11), 29-31 Oct. 2011, Porto, Portugal, Springer Verlag, Lecture Notes in Computer Science, 7014, Springer, pp. 150-161, DOI: 10.1007/978-3-642-24800-9_16, Online ISBN: 978-3-642-28765-7, [Preprint]

    [12] Fiosina, J. and Fiosins, M. (2011). Resampling Approach to Comparison of Two Routes in Stochastic Graph. In CD Proc. of the 14th Int. Conf. Applied Stochastic Models and Data Analysis (ASMDA2011), 7-10 Jun. 2011, Rome, Italy, pp. 457-464.

    Other project related publications: (not funded by ADMIT)

    [13] Andronov A. and Fiosina J., (2013). Resampling-based nonparametric statistical inferences about the distributions of order statistics. Journal of Mathematical Sciences, 191(4): 485-491, Springer US.

    [14] Fiosins, M.; Fiosina, J.; Müller, J. P. and Görmer, J.(2011). Reconciling Strategic and Tactical Decision Making in Agent-Oriented Simulation of Vehicles in Urban Traffic. In Proc. of 4th Int. ICST Conf. on Simulation Tools and Techniques (SimuTools'2011), ICST (Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering), Brussels, Belgium, pp. 144-151, ISBN: 978-1-936968-00-8, [Preprint]

    [15] Fiosins, M.; Fiosina, J.; Müller, J. and Görmer, J. (2011). Agent-Based Integrated Decision Making for Autonomous Vehicles in Urban Traffic. In Demazeau, Y. et. al. (Eds.): Advances on Practical Applications of Agents and Multiagent Systems. Springer Berlin/Heidelberg, Advances in Intelligent and Soft Computing, 88, pp. 173-178, DOI 10.1007/978-3-642-19875-5_22, Online ISBN: 978-3-642-19875-5 [Preprint]

    [16] Fiosina, J. and Fiosins, M. (2011). Statistical Estimation for Reliability Model Based on Shot-Noise Processes in a Case of Small Samples. Journal of Quality Technology and Quality Management, 8(4): 451-462.[PDF].

    Nach oben

    Training Activities/ transfer of knowledge activities/integration activities

    Dr. Fiosina has served as a reviewer of five international conferences (DoCEIS’12, AAMAS’12, WI-IAT’12, AAMAS’13, DoCEIS'14), and as a Member of Technical Committee of the Conference: 2012 IEEE Symposium on Business, Engineering & Industrial Applications.

    Since 2014 she is in Programming Committee of International Workshop on Agent-Based Distributed Data Analysis and Mining (AGENDA) at International Conference on Practical Applications of Agents and Multi-Agent Systems (PAAMS), University of Salamanca (Spain).

    In time period from March 2012 till October 2012, she has been involved in the scientific and educational coordination of the inter-university e-learning project ATLANTIS.

    Dr. Fiosina participated in two full-day post-graduate level courses organized by the NTH Operations Management & Research (OMaR) Doctoral School:

    Prof. Dr. Michael Kolonko: Stochastische Simulation (23.09.2011)

    Prof. Dr. Christoph Schwindt: Diskrete ereignisorientierte Simulation mit ExtendSim (30.09.2011)

    Dr. Fiosina improved her German language skills by attending Intensive German Language Courses at levels B1 - C1: 01.06.2011 – 16.03.2012 (540 hours)

    By successfully completing these courses, she obtained the certificate (the copy was attached), which is equivalent to C2 (professional user level) of the knowledge of German language.

    German language phonetics course (Phonetik -Sprecherziehung): 10.04.2012 – 18.07.2012 (32 hours)

    Dr. Fiosina improved her communication English language skills by taking part in a language course: General English Intermediate/ Upper Intermediate B2: 10.04.2012 – 18.07.2012 (64 hours)

    Dr. Fiosina improved her networking capabilities to foster contacts among people working in correlated areas of interest across multiple location and institutions inside and outside TU Clausthal. She made various contacts attending international conferences and using MEC Group research collaborations.

    Transfer of knowledge activities

    Starting in April 2013 Dr. Fiosina participated in teaching activities creating a very good channel to transfer the knowledge advanced during the running of the ADMIT project. Dr. Fiosina offered a module on Distributed Data Analysis (32 hours) as a part of the graduated and PhD level course ‘Distributed Data Analysis and Machine Learning’ at TU Clausthal, which is also transmitted for international students at Leibnitz Universität Hannover and Georg-August-Universität Göttingen.

    Conference participations

    Conference participations of Dr. Fiosina (funded by ADMIT):

    [ECAI’12] 20th European Conf. on Artificial Intelligence, Workshop: ‘Ubiquitous Data Mining (UDM’12)’, 27–31 Aug. 2012, Montpelier France, //Fiosina J., Fiosins M. Cooperative Kernel-Based Forecasting in Decentralized Multi-Agent Systems for Urban Traffic Networks

    [DCAI’12] 9th Int. Conf. on Distributed Computing and Artificial Intelligence, 28-30 Mar. 2012, Salamanca, Spain. //Fiosina J. Decentralised Regression Model for Intelligent Forecasting in Multi-agent Traffic Networks.

    [IDA'11] 10th Int. Sym. on Intelligent Data Analysis, 29-31 Oct. 2011,Porto, Portugal, Fiosina J. and Fiosins M. Resampling-based Change Point Estimation.

    [ASMDA’11] 14th Int. Conf. Applied Stochastic Models and Data Analysis, 7-10 Jun. 2011, Rome, Italy, //Fiosina, J. and Fiosins, M. Resampling Approach to Comparison of Two Routes in Stochastic Graph.

    Other project related conferences participations: (not funded by ADMIT)

    [HAIS’13] 7th International Conference on Hybrid Artificial Intelligence Systems, 11-13 September, 2013, Salamanca, Spain. // Fiosina, J., Fiosins, M. Density-Based Clustering in Cloud-Oriented Collaborative Multi-Agent Systems.

    [MATES’13] 11th German Conference on Multiagent System Technologies, 16-20 September, 2013, Koblenz, Germany, // Fiosina, J., Fiosins, M., Müller, J.P. Decentralised Cooperative Agent-based Clustering in Intelligent Traffic Clouds.

    [ISMiS’13] Int. Sym. on Management Intelligent Systems, 22-24 May, 2013, Salamanca, Spain. //Fiosina, J., Fiosins, M. and Müller, J. P. Mining the Traffic Cloud: Data Analysis and Optimization Strategies for Cloud-Based Cooperative Mobility Management.

    [DCAI’13]10th Int. Conf. on Distributed Computing and Artificial Intelligence, 22-24 May, 2013, Salamanca, Spain. //Fiosina, J. and Fiosins, M. Selecting the Shortest Itinerary in a Cloud-Based Distributed Mobility Network.

    [AAMAS’11, ADMI’11] 7th International Workshop on Agents and Data Mining Interaction in conjunction with the 10th Int. Conference on Autonomous Agents and Multiagent Systems, Taipei, Taiwan, 2-6 May, 2011. //Fiosins, M., Fiosina, J. and Müller, J. P. Change Point Analysis for Intelligent Agents in City Traffic.

    [SimuTools'2011] 4th Int. ICST Conf. on Simulation Tools and Techniques, 21-25 March, 2011, Barcelona, Spain. //Fiosins, M.; Fiosina, J.; Müller, J. P. and Görmer, J. Reconciling Strategic and Tactical Decision Making in Agent-Oriented Simulation of Vehicles in Urban Traffic.

    [PAAMS’2011] 9th International Conference on Practical Applications of Agents and Multi-Agent Systems, 6-8th April, 2011, Salamanca, Spain. //Fiosins, M.; Fiosina, J.; Müller, J. P. and Görmer, J Reconciling Strategic and Tactical Decision Making in Agent-Oriented Simulation of Vehicles in Urban Traffic

     

     

    Kontakt  Suche  Sitemap  Datenschutz  Impressum
    © TU Clausthal 2017