期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2018
卷号：6
期号：4
页码：3618-3623
DOI：10.15680/IJIRCCE.2018.0604068
出版社：S&S Publications
摘要：High dimensional information concerns expansive volume, mind boggling, mounting informational
indexes with various, self-governing sources. As the Data expanding radically everyday, it be a noteworthy concern to
oversee and compose the information productively. This developed the need of machine learning systems. With the
Fast advancement of Networking, information stockpiling and the information gathering limit, Machine learning bunch
calculations are presently quickly growing in all science and building spaces, for example, Pattern acknowledgment,
information mining, bioinformatics, and proposal frameworks. In order to help the adaptable machine learning system
with Map Reduce and Hadoop bolster, we are utilizing YARN Yet Another Resource Negotiator to deal with the High
Voluminous information. Different Cluster issues, for example, Cluster propensity, partition, Cluster legitimacy, and
Cluster recital canister be effectively overwhelmed by YARN bunching calculations. Mahout oversees information in
four stages i.e., bringing information, content mining, bunching, arrangement and community oriented separating. In
the proposed approach, different information writes, for example, Numbers, Raw Data and 3D-Images however,
datasets are arranged in the few classifications i.e., Collaborative Filtering, Clustering, Classification or Frequent Item
set Mining. A portion of the Pre-bunching strategies are likewise executed such lemmatization, stemming and K-Means
Hadoop based document Clustering to provide the effective and accurate information and documents on the fly vide
Big Data.
关键词：Big Data; Hadoop; Yet Another Resource Negotiator (YARN); K;Means;
