Survey of Mining Order-Preserving Submatrices from Data with Repeated Measurements

There some situations, where relative magnitude of data items have more importance than their accurate values. In those situations, Order-preserving submatrices (OPSMs) are proving to be successful in detecting concurrent patterns in data. For example relative magnitudes have importance in the process of analyzing gene expressions profiles which are extracted from microarray experiments. There are two reasons for it 1. Relative magnitudes describe changes in gene activities among whole experiment.2. In these situations, accurate values cannot be trusted since high level of noise is present. As data noise generating problems, Number of experiments has been carried out for collecting multiple measurements to address this issue. An advanced model of OPSM can be experimented and studied in which set of values obtained from replicated experiments are used to represent each data item. New problem can be called OPSM-RM (OPSM with Repeated Measurements). Depending on number of practical requirements, OPSM-RM can be characterized. Generic mining algorithm can be implemented by studying challenges of OPSM-RM.