Multi-dimensional analysis on data sets for information retrieval
Yong Shi, Arialdis Japa · 2017
With the advances in science and technology, data sets with constant changes are generated in various areas everyday. Efficient and effective approaches are in demand to retrieve valuable information from the data sets. There are several research fields in data mining which includes outlier detection that is used to discover the exceptional behaviors of certain objects. Another research field in data mining is clustering which separates data points into different groups, in a way that data points in the same group have high similarity and data points from different groups are different from each other. In this paper we redefine the meaning of outliers from a new perspective, and propose an approach to dynamically adjust the set of outliers and clusters as the data set changes. We conduct experiments to evaluate the performance of our algorithm.