Journal:Informatica
Volume 23, Issue 3 (2012), pp. 391–404
Abstract
The article describes multi-function system testing based on fusion (or revelation) of clique-like structures. The following sets are considered: (i) subsystems (system parts or units/components/modules), (ii) system functions and a subset of system components for each system function, and (iii) function clusters (some groups of system functions which are used jointly). Test procedures (as units testing) are used for each subsystem. The procedures lead to an ordinal result (states, colors) for each component (e.g., ‘out of service’, ‘major faults’, ‘minor faults’, ‘trouble free service’). For each system function a graph over corresponding system components is examined while taking into account ordinal estimates/colors of the components. Further, an integrated graph for each function cluster is considered (this graph integrates the graphs for corresponding system functions). For the integrated graph structure revelation problems are under examination (revelation of some subgraphs which can lead to system faults). Numerical examples illustrate the approach and problems.
Journal:Informatica
Volume 19, Issue 1 (2008), pp. 135–156
Abstract
Data stream mining has become a novel research topic of growing interest in knowledge discovery. Most proposed algorithms for data stream mining assume that each data block is basically a random sample from a stationary distribution, but many databases available violate this assumption. That is, the class of an instance may change over time, known as concept drift. In this paper, we propose a Sensitive Concept Drift Probing Decision Tree algorithm (SCRIPT), which is based on the statistical X2 test, to handle the concept drift problem on data streams. Compared with the proposed methods, the advantages of SCRIPT include: a) it can avoid unnecessary system cost for stable data streams; b) it can immediately and efficiently corrects original classifier while data streams are instable; c) it is more suitable to the applications in which a sensitive detection of concept drift is required.