Association rule mining activity : A general formalism

Mining association rules from a large collection of databases is based on two main tasks. One is generation of large itemsets; and the other is finding associations between the discovered large itemsets. Existing formalism for association rules are based on a single transaction database which is not sufficient to describe the association rules based on multiple database environment. In this paper, we give a general characterization of association rules and also give a framework for knowledge-based mining of multiple databases for association rules.

DNA-binding and nuclease activity of 1,10-phenanthroline-based thiocyanato, acetato, azido and benzoato bridged dinuclear copper(II) complexes

The mononuclear Cu(II) complex [Cu(phen)(H2O)(NO3)(2)] (1), obtained by the reaction of 1,10-phenanthroline with Cu(NO3)(2)center dot 3H(2)O in methanol solution, reacts with anionic ligands SCN-, AcO-, N-3(-) and PhCO2- in MeOH solution to form the stable binuclear complexes [Cu-2(H2O)(2)(phen)(2)(mu-X)(2)](2) (NO3)(2), where X = SCN- (2), AcO- (3), N-3(-) (4) or PhCO2- (5). The molecular structure of complex 3 was determined by single-crystal X-ray diffraction studies. These complexes were characterized by electronic, IR, ESR, magnetic moments and conductivity measurements. The electrochemical behaviour of the complexes was investigated by cyclic voltammetry. The interactions of these complexes with calf thymus DNA have been investigated using absorption spectrophotometry. Their DNA cleavage activity was studied on double-stranded pBR322 plasmid DNA using gel electrophoresis experiments in the absence and presence of H2O2 as oxidant.