Ji Sun Sally Kim IWA #3

docx

School

Harrisburg University of Science and Technology *

*We aren’t endorsed by this school

Course

500

Subject

Health Science

Date

Dec 6, 2023

Type

docx

Pages

Uploaded by JudgeEchidnaMaster1024

Assignment #3 Ji Sun “Sally” Kim Harrisburg University of Science and Technology HCIN 500-50 A Dr. Steven Hardy September 12, 2023

1 1. Discuss the 3 major groups of analytic approaches in terms of the kinds of questions answered by their output The first analytical approach is descriptive. Descriptive analytics processes data to look for patterns and reports the current situation or problem (Hersh, 2018, p. 420; Hoyt et al., 2018, 1276—1277). For example, a system with every patient’s health information might report that there were nine cases of lead poisoning in a certain region or observe a pattern amongst patients that patients with BMI above 25 are associated with type 2 diabetes with a certain percentage level of confidence. A notable element of descriptive analytics is that the user does not set a target value or class as the system itself is conducting unsupervised learning (Hoyt et al., 2018, p. 1277). The second type of analytics is predictive analytics, which uses “simulation and modeling techniques that identify trends and portend outcomes of actions taken” (Hersh, 2018, p. 420). Unlike descriptive analytics, predictive analytics has a target variable that the user sets, which means predictive analytics is based on supervised learning (Hoyt et al., 2018, p. 1279). Predictive analytics could plot a linear model that shows the weight of a newborn baby (dependent variable) in relation to the age of the mother (independent variable), which could help the user attain a formula to estimate the weight of a baby and potentially prepare for a baby with high health risks based on the mother’s age before the actual delivery date. Predictive analytics could also create a non-linear model that categorizes the given data into clusters (Hoyt et al., 2018, p. 1281). The third and last major group of analytic approaches is prescriptive analytics, which uses descriptive and predictive analytics to generate a solution or an actionable item to optimize future outcomes based on the current situation (Hersh, 2018, p. 420). A simple example of

2 prescriptive analytics is the suggestions one receives when typing out a document, an email, or a text message to finish the rest of the word or the sentence. Based on all of the past learnings, the system assesses the user’s current situation and prescribes a solution or an otherwise actionable item. This could be something like the system suggesting to the clinician to prescribe the patient a certain treatment or medication over a different option based on the things the clinician was typing into the health information database. 2. What operational criteria are commonly used to define the label "Big Data"? “Big Data” has five attributes of that can help serve as operational criteria to help define it. First, there must be large volumes of data that “are being generated each minute” that “it can’t be analyzed or stored on one computational unit” (Hersh, 2018, p. 421; Hoyt et al., 2018, p. 30). The second factor is based on the velocity, as in whether the data is “being generated so rapidly that it needs to be analyzed without placing it in a database” (Hersh, 2018, p. 421; Hoyt et al., 2018, p. 30). The third criterion is whether there is such a tremendous variety of data that “roughly 80% of data in existence is unstructured so it won’t fit into a database or spreadsheet” (Hersh, 2018, p. 421; Hoyt et al., 2018, p. 30). The fourth factor is based on the veracity of the data. The “missing data and other challenges” in the current data are not as significant of a problem as in the past as they are flooded out by the because of the massive amounts of data from credible sources that are also being generated (Hersh, 2018, p. 421; Hoyt et al., 2018, p. 30). The fifth and last criterion is value. The large volumes of unstructured data would not be meaningful or useful to data scientists without having value to it (Hersh, 2018, p. 421; Hoyt et al., 2018, p. 30).

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

3 References Hersh, W. R. (2018). Healthcare Data Analytics. In Health Informatics: Practical Guide (7th ed., pp. 420). Lulu.com. Retrieved September 1, 2023, from https://www.scribd.com/read/485889587/Health-Informatics-Practical-Guide-Seventh- Edition. Hoyt, R. E., Bernstam, E. V., & Hersh, W. R. (2018). Overview of Health Informatics. In Hoyt, R. E., & Hersh, W. R. (Eds.), Health Informatics: Practical Guide (7th ed., p. 30). Lulu.com. Retrieved September 1, 2023, from https://www.scribd.com/read/485889587/Health-Informatics-Practical-Guide-Seventh- Edition. Hoyt, R. E., Snider, D., & Mantravadi, S. (2018). Introduction to Data Science. In Hoyt, R. E., & Hersh, W. R. (Eds.), Health Informatics: Practical Guide (7th ed., pp. 1276—1281). Lulu.com. Retrieved September 1, 2023, from https://www.scribd.com/read/485889587/Health-Informatics-Practical-Guide-Seventh- Edition.

Ji Sun Sally Kim IWA #3

Related Documents