Introducing a new way of noise-clustering big data

Introducing a new way of noise-clustering big data

Mehdi Nezhadfarhani1 Hossein Chahkhouei2 Amir Mehdi Rajabi3 Mohammad Hossein Jamshisi4

1) PhD student in Computer Software Engineering
2) PhD student in Computer Software Engineering Email:

IOI Code : XBHB-EBBAK
Publication : International Congress of Science and Engineering - TOKYO UNIVERSITY - JAPAN(tuicet.com)
Abstract :
    Clustering of big data based on similarity is one of the most important steps in data analysis and is one of the most widely used data mining tools. Clustering is a difficult problem in the field of big data. This difficulty is due to the large amount of data received from a stream. In addition to the large amount of data input data can cause noise, which can cause clustering problems. In this study, a density-based clustering method is presented that can perform clustering operations in parallel. In this method, a neural network with a training dataset is used for noise detection and this dataset is used for noise detection. The proposed method is tested on a large data of traffic accidents and the results show that the proposed method performs well.

Keywords : bulk data clustering data mining network