Libraries at University of Nebraska-Lincoln



Unstructured information is continuously irregular and streaming information from such a sequence is tedious because it lacks labels and accumulates with time. This is possible using Incremental Clustering algorithms that use previously learned information to accommodate new data and avoid retraining. This paper therefore seeks to understand the status of "Distributed Incremental Clustering" on images with text and numerical values, its limitations, scope, and other details to devise a better algorithm in future. To further enhance the analysis, we have also included methodology, which can be used to perform clustering on images or documents based on its content.