@Article{jihpp.2020.010223,
AUTHOR = {Dunhong Yao, Yu Chen},
TITLE = {Design and Implementation of Log Data Analysis Management System Based  on Hadoop},
JOURNAL = {Journal of Information Hiding and Privacy Protection},
VOLUME = {2},
YEAR = {2020},
NUMBER = {2},
PAGES = {59--65},
URL = {http://www.techscience.com/jihpp/v2n2/40535},
ISSN = {2637-4226},
ABSTRACT = {With the rapid development of the Internet, many enterprises have 
launched their network platforms. When users browse, search, and click the 
products of these platforms, most platforms will keep records of these network 
behaviors, these records are often heterogeneous, and it is called log data. To 
effectively to analyze and manage these heterogeneous log data, so that 
enterprises can grasp the behavior characteristics of their platform users in time, 
to realize targeted recommendation of users, increase the sales volume of 
enterprises’ products, and accelerate the development of enterprises. Firstly, we 
follow the process of big data collection, storage, analysis, and visualization to 
design the system, then, we adopt HDFS storage technology, Yarn resource 
management technology, and gink load balancing technology to build a Hadoop 
cluster to process the log data, and adopt MapReduce processing technology and 
data warehouse hive technology analyze the log data to obtain the results. Finally, 
the obtained results are displayed visually, and a log data analysis system is 
successfully constructed. It has been proved by practice that the system 
effectively realizes the collection, analysis and visualization of log data, and can 
accurately realize the recommendation of products by enterprises. The system is 
stable and effective.},
DOI = {10.32604/jihpp.2020.010223}
}