Page 242 - Big Data Analytics for Intelligent Healthcare Management

9.3 HEALTH RECOMMENDATION SYSTEM 235

Coverage deals with the percentage of items that the recommendation system was able to recommend.
If there are many sparse user rated items, it is not possible to compute prediction. It can be minimized
by defining small neighborhood sizes for user or items. By considering the above parameter measures,
we can evaluate any type of recommendation system so that performance can be achieved in term of
accuracy.

9.3 HEALTH RECOMMENDATION SYSTEM
With the rapid development of data mining and analytics, there is a rise in the application of big data
analytics in various domains. The healthcare system has emerged as one of the promising sectors in
which big data analytics and its application has earned their own recognition and honors. The three
main distinguishing characteristics of big data found in healthcare data are Volume (the amount of data
produced by organizations or individuals. The sources may be internal or external), Velocity (the rate at
which data is generated, captured, and shared), and Variety (presence of data from wide range of
sources with different formats). Besides these three Vs, there is another V known as veracity (whether
the obtained data is correct or consistent), which plays an important role in healthcare. With huge quan-
tities of unprocessed data and information overload, recommendation systems are becoming popular
for their role in filtering large datasets and information. There is a need for new health recommendation
systems (HRSs) that can improve the healthcare system and handle information from multiple patients
suffering from different diseases at one time [21–23].
The recommendation system is based on predictive analytics, which predicts and recommends ap-
propriate items to the users. This system can be applied to specific applications. Healthcare analytics is
a major area in big data analytics, which can be incorporated into a recommendation system. The
health-based recommendation system is a decision-making system that recommends proper healthcare
information to both health professionals and patients as end users. By using this system, patients are
recommended to apply proper treatment of disease to avoid health risks and health professionals are
able to retrieve valuable information for clinical guidelines along with delivering high-quality health
remedies to the patients. This HRS should be trustworthy and reliable so that end users can use this
system for their benefits [12, 24, 25] (Fig. 9.4).
This HRS consists of different phases through which a particular item is recommended. These are
training phase, user profile process phase, sentimental analysis phase, privacy preservation phase, and
recommendation phase. First, we have to collect the healthcare dataset on which we apply the feature
selection and classification method. A major part of this HRS is to collect and prepare profile health
records (PHR) and a user database. PHR is very important as input for the recommendation engine to
predict and recommend health remedies to the patients. We collect useful information from the user
database, which is connected to PHR, for feature selection. Then, we apply a classification algorithm to
classify and store in the knowledge repository. The recommendation process internally uses three sub-
phases that is, the collection phase, the learning phase, and the recommendation phase. After applying
these phases of the recommendation process by using the user database, proper treatments are recom-
mended to patients and health professionals are recommended to use valuable clinical guidelines and
high-quality healthcare treatment.