A rough set approach to multiple dataset analysis

Abstract

In the area of data mining, the discovery of valuable changes and connections (e.g., causality) from multiple data sets has been recognized as an important issue. This issue essentially differs from finding statistical associations in a single data set because it is complicated by the different data behaviors and relationships across multiple data sets. Using rough set theory, this paper proposes a change and connection mining algorithm for discovering a time delay between the quantitative changes in the data of two temporal information systems and for generating the association rules of changes from their connected decision table. We establish evaluation criteria for the connectedness of two temporal information systems with varying time delays by calculating weight-based accuracy and coverage of the association rules of changes, adjusted by a fuzzy membership function.

Notice of Retraction: Research on limitation of knowledge

“…[1] The activities of the static knowledge are produce, remember and recall and represent. [2] So no matter how hard we try, we'll miss something compared with the real world with no doubt when we create the static knowledge. We can't use this incomplete mirror to rebuild the real object by all the relevant static knowledge.…”

An advancing investigation on reduct and consistency for decision tables in Variable Precision Rough Set models

“…It provides a formal methodology aiming at data analysis problems that involve uncertain, imprecise or incomplete information, and has had widespread success in many artificial intelligence research fields [4], [5], [6]. However, when the given information contains some errors, such as missing information and classification abnormalities or the given Decision Table (DT) is derived from a relatively smaller data set, the obtained results of the classical RS model cannot always perform well and shows a poor generalization ability [7], [8].…”

Meta-association rules for mining interesting associations in multiple datasets

“…changes, connections) from multiple datasets. When the temporal dimension comes into play, there exists approaches like the one developed in [24], for discovering changes and connections in two temporal information systems by employing different types of association rules using rough set theory. However, our approach is intended for the discovery of the existent commonalities from multiple datasets where a set of association rules have been already discovered, obtaining a set of meta-association rules which contains the more frequent rules in the majority of datasets.…”