Sonar signals recognition is an important task in detecting the presence of some significant objects under the sea. In military sonar signals are used in lieu of visuals to navigate underwater and/or locating enemy submarines in proximity. Specifically, classification in data mining is useful in sonar signal recognition in distinguishing the type of surface from which the sonar waves are bounced. Classification algorithms in traditional data mining approach offer fair accuracy by training a classification model with the full dataset, in batches. It is well known that sonar signals are continuous and they are collected in streaming manner. Although the earlier classification algorithms are effective for traditional batch training, it may not be practical for incremental classifier learning. Because sonar signal data streams can amount to infinity, the data pre-processing time must be kept to a minimum to fulfill the need for high speed. This paper introduces an alternative data mining strategy suitable for the progressive purging of noisy data via fast conflict analysis from the training dataset without the need to learn from the whole dataset at one time. Simulation experiments are conducted and superior results are observed in supporting the efficacy of the methodology.