Page 247 - Big Data Analytics for Intelligent Healthcare Management

240 CHAPTER 9 INTELLIGENCE-BASED HEALTH RECOMMENDATION SYSTEM

undisturbed and hassle-free manner to doctors so that doctors can interpret the symptoms of patients
and can give recommendations for a particular disease to patients. The visualization of data should
address the purpose of the recommendation system and be able to understand the patients, doctors,
and their intentions. There should be a proper visualization tool in a recommendation system that fos-
ters the user’s willingness to explore options and help explain individual recommendations. Since in-
dividual differences might play a vital role in the healthcare sector, it is crucial to intensify research in
this field.
Some of the common big data tools that are used in the healthcare sector are DataCleaner, Apache
Hadoop, Cassandra database etc. Apache Hadoop is the most prominent tool in the big data industry
with its enormous capability of large-scale processing data. Hive is also one of the eco-components of
Hadoop, which allows programmers to analyze large datasets on the Hadoop platform. It helps with
querying and managing large datasets. DataCleaner is a data quality analysis platform, which has a
strong data profiling engine. This tool is usually used for data cleaning, data transformation, and data
merging. Today’s Cassandra database is widely used to effectively manage a large amount of data.
These tools are used to work with recommendation engines in big data analytics [33–35].

9.3.4 EVALUATION OF HRS
For the success of the recommendation system, it is very important to choose what type of criteria to
evaluate the system with. Conventionally, recommendation systems were evaluated based on criteria
borrowed from information retrieval [36]. Common metrics used in the evaluation are:
i. Precision: The measure of retrieved instances that are relevant.
ii. Recall: The fraction of correctly recommended items that are also part of the collection of useful
recommended items.
iii. F-Measure: It is a measure of a test’s accuracy and is defined as the weighted harmonic mean of the
precision and recall of the test.
iv. ROC-curve: ROC-curve is a way to compare diagnostic tests. It is a plot of the true positive rate
against the false positive rate. It is used to represent the relationship between sensitivity and
specificity.
v. RSME: This measure defines standard deviation of the residual errors that is, differences between
predicted values and known values.

Evaluation criteria of a recommendation system are needed to measure the strength of the HRS based
on user acceptance and satisfaction. By making the system suitable for individual users, the system can
run as per users’ requirements so that user will not face any problems and this ultimately leads to better
medical research. This includes user diversity research, not just in regard to user-specified results, but
also in regard to the user interface of the HRS. The pretentiousness of accuracy metrics and underrep-
resentation of metrics such as serendipity and coverage pose a serious challenge in a classic recom-
mendation system. Rare diseases are very uncommon but the collection of data and case studies of
similar cases can help. Therefore, finding all relevant results is important for the HRS. Another vital
research issue is trust. If things worsen, the doctor can program the system to take actions so that trust is
maintained.
While designing HRSs, the person concerned should be careful and prepare a plan according to the
requirements. The capability of a recommendation system can be appraised in regard to the user’s
external behavior. The measure of the effectiveness of a HRS depends on behavioral evaluations.