Question

DisCUSSION QUESTION 4. When discussing KDDM, what is the meaning of the fol- analytics questions you would want to consider in your work setting. lowing adage: "Garbage in, garbage out"? 5. Fig. 23.7 includes 10 components every data governance program should employ. For each component, provide one implication of that component within a healthcare setting. , Compare and contrast the types and methods that are available for EDA, predictive, and prescriptive analytics. Give examples of each. s Why is it necessary to process coded and text data before using data mining methods?

Answer 1

1. Big Data is a collection of data sources inside and outside your company that represents a source for ongoing discovery and analysis.

4 Vs of Big Data:

• Velocity (speed of the data)
• Volume (Scale of the data. Increase in the amount of data stored.
• Variety (Diversity of the data.)
• Veracity (Certainty of the data. Accuracy)

Companies like Amazon and Netflix use algorithms based on big data to make specific recommendations based on customer preferences. When Amazon uses recommendations for you to purchase something. Recommendations engines are a common application of big data.

Siri also uses big data to devise answers to the infinite number of questions and end users may ask.

According to McKinsey in 2013, the emergence of cloud computing has highly contributed to the launch of the Big Data era.

When we look at big data, we can start with a few broad topics:

• integration
• analysis
• visualization
• optimization
• security
• governance.

Programming is an essential big data analysis skill. What makes it extra special, though, is the versatility. You can, and must, learn multiple technologies that will help you grow as a Big Data analyst.

But, technologies are not limited to programming alone. The range of technologies that a good big data analyst must be familiar with is huge. It spans myriad tools, platforms, hardware, and software. For example, Microsoft Excel, SQL, and R are basic tools. At the enterprise level, SPSS, Cognos, SAS, MATLAB are important to learn as are Python, Scala, Linux, Hadoop, and HIVE.

The actual technologies that you use will depend upon the environment you are working in. It will also vary based on the requirements of your company and project.

Challenges to Data Analytics:

Data generated in the routine care of patients may be limited in its use for analytical purposes.

1. incompletely adhere to know standards

2. issues of cause & effect

3. what can be answered, as opposed to prospective hypotheses

4. bigger not always better

5. ethical concerns

6. who can use it.

2*.Descriptive Analytics, which use data aggregation and data mining to provide insight into the past and answer: “What has happened?”

Summarize what has happened in the past and allow decision-makers to learn from past behaviors
BA tools in descriptive analytics applications include online analytical processing, data mining, decision support systems, and a variety of statistical procedures

Standard types of reporting that describe current situations and problems

eg: how many uninsured patients with type 2 diabetes

*Predictive Analytics, which use statistical models and forecasts techniques to understand the future and answer: “What could happen?”

Examine recent and historical data to detect patterns and predict future outcomes and trends
Predictive analytics provide estimates about the likelihood of a future outcome.

Simulation and modeling techniques that identify trends and portend outcomes of actions taken

Eg: predict who will be readmitted for heart failure

*Prescriptive Analytics, which use optimization and simulation algorithms to advice on possible outcomes and answer: “What should we do?”

Go beyond descriptive and predictive models by recommending one or more courses of action and showing the likely outcome of each decision
Prescriptive analytics attempt to quantify the effect of future decisions in order to advise on possible outcomes before the decisions are actually made
Prescriptive analytics require predictive analytics with two additional components: actionable data and a feedback system that tracks the outcome produced by the action taken

Prescriptive analytics use a combination of techniques and tools such as business rules, algorithms, machine learning, and computational modeling procedures. These techniques are applied against input from many different data sets including historical and transactional data, real-time data feeds, and big data.

Optimizing clinical, financial, and other outcomes

Eg: cost-effective to case manage in the hospital or home

3. Data mining is the processing and modeling of large amounts of data to discover previously unknown patterns or relationships

Codesets have been used for dealing with a vast range of different size value sets for a considerable range of different purposes. Most of these codesets are local to a geographical region, local to individual manufacturers of clinical systems, or nationally mandated statistical data collections. The proliferation of codesets and the variability in the complexity of codesets combine in a way that inhibits semantic interoperability if they need to co-exist in a given clinical system.

As a general principle, codes are for computers, not humans. Codes should work behind the scenes and not be exposed to users, particularly busy clinicians. They should not be deliberately exposed, unless absolutely necessary, to those who are only peripherally likely to understand their meaning, such as software developers, or data modelers. Writing standards and specifications for humans, that are littered with abbreviations and codes often dreamed up on a whim, that has to be understood, transcribed, embedded in the program code, put into test scripts and test specifications and otherwise discussed and manipulated, and above all remembered, is fraught with danger. It is not sound engineering practice. It dramatically narrows the pool of experts who can understand and use the specifications, and risks misunderstanding and transcription errors and the resultant clinical errors that can ensue.

In health IT, examples might be genes and gene sequences, tumor staging, or the classification of diseases. In these circumstances, it is often easier for the humans involved to refer to these concepts by codes. Where codes are to be used by humans, it is sensible for the codes to carry additional meaning or representational hints to aid the humans to disambiguate the codes and reduce the chance of error during human processing and transcription.

EHR systems impose requirements on data far exceeding those required in messages. Data may come from a vast array of sources, including direct input by humans, messages from laboratories, pharmacies etc., referral and other documents from other healthcare providers, etc.

Eg:

Code	Meaning
M	Male
F	Female
U	Undifferentiated

4. Knowledge Discovery in Data Mining(KDDM) facilitates this transformation and can be applied in any business regardless of size or location. KDDM is the process ofexploring datato discover previously undetected patterns whichcan then be leveraged into knowledge andtherefore facilitatebusiness intelligence.Datamining is only one step in this process, where thealgorithms are used to explore the prepared data.

The KDDM process includes an iterative set of steps to help bring value to data through the discovery and application of knowledge in the business, and one of the processes is prepared data.

Prepare the data

. The data is prepared, cleaned and preprocessed to ensure that business objectives and data quality standards can be attained. The old adage of garbage in, garbage out is applicable here.