Range-support Association Pattern (RAP) is an novel approach based on association pattern analysis for discovering constant-row biclusters in gene expression data. Contrary to traditional association pattern discovery approaches, RAP works with real valued data sets without discritizing them. RAP discovers small highly coherent biclusters as opposed to large blocks discovered by traditional biclustering approaches. Data sets and code are available here.


SMK aims at the efficient discovery of discriminative patterns from biological data with high density and high dimensionality (e.g. Gene Expression data, and SNP data), and especially for the discovery of those patterns with relatively low-support but high discriminative power (e.g. odds ratio, information gain, p-value etc), which complements existing discriminative pattern mining algorithms. Data sets and code are available here.


Traditional association mining algorithms use a strict definition of support that requires every item in a frequent itemset to occur in each supporting transaction. In real-life data sets, this limits the recovery of frequent patterns as they are fragmented due to random noise and other errors in the data. We implemented a suite of algorithms to discover approximate frequent itemsets in the presence of noise. Source code can be downloaded from here.