Reduce Data Anomalies Using Manifold Learning

Manifold learning has recently appeared as a powerful method for dimensionality reduction. Most studies and theoretical results in the field of this method have only focussedon preserves distances quite nicely; however, empirical results are sparse. In this paper we select the important features of dataand assignment rank for high value data and the penalty for low value data or similar data, then they insert into manifold learningalgorithm LSML. Next, the general method of dealing with bothnormal data and anomal data is discussed. If the anomalies occur on low value data, they are removing with dimantional reductionbut if anomalies occured on high value data to retrieve them .The propose Error function to be divided by the distance between thenormal data point and anomaly data point and add data penalty, it will help remove. The methode provides a way to map anumber of points in high dimensional spasce into a low dimentional space, with only smal distortion of the distancees between the points.