Extraction of Association Rules from Tobacco Smoke Effect on the Placenta Microarray Dataset using Gene Ontology Based Optmized Association Rule Mining

Gene Ontology (GO) is said to be the most popular bio-ontology that describes the gene products and its characteristics. For describing the gene product and its characteristics the three terms are used, namely, biological process, molecular function and cellular component. GO annotation is the term that is derived from kind of sub-ontologies at various stages. It is a vital source that describes the relationship between the three sub-ontologies. For effective information finding the data mining approach names as association rule mining, which is modified for mining the relationships from various ontologies from GO annotation data. For mining the relationship the abstractions are mandatory. In the existing system, GO-WAR (GeneeOntology-based Weighted Association Rules) methodology was proposed by using a FP - growth algorithm for extracting weighted association rules. GO-WAR is could extract association rules with a high information(IC)without loss of support and confidence from a dataset of annotated data. However, the performance was low for extracting weighted association rules. This paper introduces the new methodology GOPAR (Gene Ontology based Predictive Association Rule) to eliminate the drawback of GO-WAR. GOPAR avoids the repeated association rule and generating the predictive grouped rules, the grouping of association rules strictly eliminates multiple GO terms. Further, the paper presents the GOPOAR (Gene Ontology based Predictive Optimized Association Rule) approach is to achieve a best optimal solution and find the missing values of predictive association rule. Based on the experimental result the GOPOAR extracted more number of significant rules also providing truthful and better optimal results.