Application of Ontological Modelling Methods and Text Classification Algorithms for Storage System Faults Detection

Abstract

This paper describes application of diagnostic model, created with ontological modelling methods and machine learning text classifi cation algorithms, for fault detection, based on system log messages data, in enterprise-level storage system. Proposed fault detection model uses external procedures for the description ofthe relations between parameters and states of storage systems, based on the implementation of the machine learning algorithms. As an example of such relation, author describes application of the text classifi cation method for the task of software log analysis.