Explanation-based learning to recognize network malfunctions

Several network troubles and/or malfunctions may occur due to the heavy traffic of recent computer networks. The discovering of some types of these troubles is not straightforward. Therefore, there is a real need to an intelligent system to recognize that type of problems using a priori background knowledge. The aim of this work is to present a network-monitoring utility that can discover various operational patterns and can provide sensible advice that may support the network administrator. It presents a machine learning system that can recognize network malfunctions. Such recognition process may be expressed in structured patterns to support network administrator for both problem solving and network management. To achieve this objective an explanation_based learning (EBL) procedure is used to obtain operational rules. In this case, the domain (network) knowledge is formally expressed and only one training example is analyzed in terms of this knowledge. This system uses a relational database to store and maintain the knowledge_base. The main contribution of the proposed system is to discover the abnormal patterns (malfunctions) of the network traffic. These abnormal patterns, as such, could be recognized from a real network using EBL. If the network administrator is advised with that malfunctions then he can adapt the current configuration in order to avoid the corresponding problems.

[1]  Tom M. Mitchell,et al.  Explanation-Based Generalization: A Unifying View , 1986, Machine Learning.

[2]  Stephen Kwek,et al.  Learning from examples with unspecified attribute values , 2003, Inf. Comput..

[3]  László Monostori,et al.  Artificial neural networks in intelligent manufacturing , 1992 .

[4]  Makoto Haraguchi,et al.  Data abstractions for decision tree induction , 2003, Theor. Comput. Sci..

[5]  Kornel Terplan.  Effective management of local area networks (2nd ed.): functions, instruments, and people , 1992 .

[6]  László Monostori,et al.  AI and machine learning techniques for managing complexity, changes and uncertainties in manufacturing , 2003 .

[7]  R. Michalski,et al.  Multistrategy Constructive Induction: AQ17-MCI , 1993 .

[8]  Ashwin Ram,et al.  Using Introspective Reasoning to Select Learning Strategies , 1991 .

[9]  Matthew G. Naugle.  Network Protocol Handbook , 1998 .

[10]  Michael M. Cox.  An Empirical Study of Computational Introspection: Evaluating Introspective Multistrategy Learning in the Meta-AQUA System , 1996 .

[11]  Yechiam Yemini,et al.  Managing Communication Networks by Monitoring Databases , 1991, IEEE Trans. Software Eng..

[12]  Matthew L. Hess,et al.  Multiprotocol Networking - A Blueprint , 1995, IBM Syst. J..

[13]  Jin-Wook Chung,et al.  A study on the classified model and the agent collaboration model for network configuration fault management , 2003, Knowl. Based Syst..

[14]  Catherine M. Burns,et al.  Ecological interface design: a new approach for visualizing network management , 2003, Comput. Networks.

[15]  László Monostori,et al.  Multistrategy learning approaches to generate and tune fuzzy control structures and their application in manufacturing , 1998, J. Intell. Manuf..

[16]  José-Luis Marzo,et al.  A distributed simulator for network resource management investigation , 2003, Comput. Commun..

[17]  Michael Borth,et al.  Learning from Multiple Bayesian Networks for the Revision and Refinement of Expert Systems , 2002, KI.

[18]  Sebastian Abeck,et al.  Integrated Management of Networked Systems: Concepts, Architectures and their Operational Application , 1999 .

[19]  Aijun An.  Learning classification rules from data , 2003 .

[20]  Jay Smith,et al.  Relational Database Management System (RDBMS) Management Information Base (MIB) using SMIv2 , 1994, RFC.

[21]  Rüdiger Wirth,et al.  When Distribution is Part of the Semantics: A New Problem Class for Distributed Knowledge Discovery , 2001 .

[22]  J.W. Sheppard,et al.  Explanation-based learning with diagnostic models , 1992, Conference Record AUTOTESTCON '92: The IEEE Systems Readiness Technology Conference.

[23]  Gerald DeJong,et al.  Iterated Phantom Induction: A Knowledge-Based Approach to Learning Control , 2001, Machine Learning.

[24]  Kenneth J. Christensen,et al.  Local Area Networks - Evolving from Shared to Switched Access , 1995, IBM Syst. J..