Page 255 - Big Data Analytics for Intelligent Healthcare Management

248 CHAPTER 10 COMPUTATIONAL BIOLOGY APPROACH ON GENETIC
DISORDER

overall network from different sources of attack, which help to create the straightforward way to
identify the corrected field and make suitable decisions for identification from larger databases.
Sahani et al. [5] discussed the details of classification of intrusion detection using data mining tech-
niques. In the US healthcare system, biomedical data is expected to reach the zettabyte scale [1] from
a variety of fields such as scientific instruments, electronic health records, and clinical decision sup-
port systems [6, 7]. Dey and coworker [8] discussed the advancements and innovations in the area of
medical image and data processing that have led to creation of a secure mechanism to move images
and signals over the internet. They also discussed image processing that used intelligent techniques
for medical data security.
Besides other sources, there is currently an open source data processing platform called Hadoop.
It is capable of processing a large amount of data, which allocates portions of the datasets to servers,
each of which resolve the various parts of bigger problems [9]. It allows the users to exploit com-
putation through mapreduce implementation. Depending on the progress of better performance,
Hadoop breaks up a file into different blocks and keeps them in various nodes within a cluster. Reddy
et al. [10] provided details regarding a data aware scheme for scheduling big data application with
SAVANNA Hadoop. In addition, CouchDB and MongoDB are also used as data analytics platforms
to aggregate data in unique ways [11]. Mapreduce is a minimization technique that generates file
indexing with sorting and mapping [12]. To regulate and handle analysis of a large amount of data
in a distributing computing manner, different data analysis tools exist, including the extract trans-
form load (ETL) process, querying, data mining [13], and online analytical processing (OLAP). Mis-
hra et al. [14] studied the foundation, applications, and challenges of cloud computing. It presents the
up-to-date computing paradigms. It plays an important role in the area of virtualization, security, and
allocation of resources to monitoring of complex optimization problems. Although data analysis has
rapidly increased insight into many aspects, there are still some ambiguities present. Cloud comput-
ing is a computing system that is dynamically scalable, which depends on parallel computers. It is a
new solution compared to the old version, as well as traditional data analysis problem. Sarkhel and
coworker [15] proposed different task scheduling algorithms such as minimum-level priority queue
(MLPQ), Min-Median, and Mean-MIN-MAX to reduce the makespan with maximum utilization of
the cloud. Distributed data mining (DDM), based on typical software-as-a-service, also works for
fulfillment of demanding data processing. It can scrutinize the excess data basedonparallelcom-
puting and storage capacity. This DDM revamps the user’s demand in the workflow process, which
can be set aside as parallel task sequences in terms of cloud computing platforms [16].Beheraand
coworker [17] discussed particle swarm optimization that is based on the metaheuristic evolutionary
optimization technique. It is quite attractive in the swarm intelligence community because of its un-
derstandable algorithm structure.

10.2 APPLICATION OF BIG DATA ANALYSIS
Big data analysis provides better policies not only to the government sector but also to industrial sec-
tors, including the manufacturing sector, agriculture sector, and healthcare sector. In the healthcare
sector, big data are not only managed by volume but it also depends on the type of data and the speed
[18]. Moreover, to boost the flow rate of the healthcare sector, electronic medical records (EMRs) are
playing a major role in the field of genomic and clinical data [19]. Data analytics may be used to deal