As previously reported, at a meeting of the 13th IEEE International Conference on the Application of Information and Communication Technologies (AICT 2019) on high-performance computing and machine learning held recently in Baku, a report was heard on Consensus clustering by weight optimization of input partitions Consensus clustering by weight optimization of input partitions” co-authored by the vice president of ANAS, director of the Institute of Information Technology, academician Rasim Alguliyev, head of department , corresponding member of ANAS, doctor of technical sciences Ramiz Aliguliev and senior researcher, Ph.D. in technology, associate professor Lyudmila Sukhostat.
The scientific reports presented at the conference were published in the “Materials of the 13th IEEE International Conference on the Application of Information and Communication Technologies”.
It should be noted that the article widely uses the resolved approach to improving the accuracy and stability of clustering results. The weighted consensus clustering approach is to master individual weight clustering methods using the purity utility function.
Consensus clustering is a promising approach to finding cluster structures from a dataset. Consensus clustering can help you find reliable units, identify noise and deviations, and combine solutions from many sources.
The article proposes a consensus clustering algorithm consisting of DBSCAN (Density-Based Spatial Clustering of Applications with Noise), OPTICS (Ordering Points to Identify the Clustering Structure), CLARANCE (Clustering Large Applications with Randomized Search), c-means and general clustering algorithms neighbors (shared nearest neighbor clustering, SNNC).
© All rights reserved. Citing to www.ict.az is necessary upon using news