AUTHOR=Yan An , Wang Wei , Ren Yi , Geng HongWei 

TITLE=A Clustering Algorithm for Multi-Modal Heterogeneous Big Data With Abnormal Data

JOURNAL=Frontiers in Neurorobotics

VOLUME=Volume 15 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/neurorobotics/articles/10.3389/fnbot.2021.680613

DOI=10.3389/fnbot.2021.680613

ISSN=1662-5218

ABSTRACT=In order to solve the problem of data abnormalities in traditional multi-modal heterogeneous big data detection algorithms and missing data, which leads to data modal confusion, a multi-view heterogeneous big data clustering algorithm based on Kmeans clustering is established. With heterogeneous data as the supporting background of big data, data analysis with the help of multiple views, determination of similarity detection metrics, and a multi-view heterogeneous system based on Kmeans. Then, use the BP neural network to predict the missing attribute values, complete the missing data, restore the big data structure in a heterogeneous state, and then use the BP neural network to denoise the abnormal data to achieve multi-view heterogeneity Research on big data detection algorithms. Both theoretical verification and experimental results show that the accuracy of the proposed method is much higher than that of the original algorithm.