Difference between revisions of "Predictive Analytics"

Revision as of 13:33, 9 April 2021

Predictive analytics is the use of algorithms and machine learning techniques to forecast future events in real-time. Leveraging vast data sets, these algorithms are able to predict potential risks and costs, or even an individual’s future behaviour.^[1] Thanks to rapid advancements in technology and the emergence of big data, the use of predictive analytics has grown in various industries, from providing healthcare treatment recommendations to assessing and determining candidates for hire and even to helping law enforcement anticipate potential crimes and criminals.^[2][3][4] As technology continues to improve, data analytics and artificial intelligence continue to grow their capabilities and expand their applications. However, with predictive analytics becoming increasingly prevalent in decision-making processes that have direct and potentially life-changing impacts on people’s lives, ethical concerns regarding algorithmic bias, transparency, and data privacy are revealed.^[5]

Types of Predictive Analytics

Questions and what form of predictive analytics can answer them.

There are three types (or models) of predictive analytics. There are predictive models, descriptive models, and decision models ^[6]. Each model has a specific function related to predicting and forecasting information. The most common type is predictive models, but other forms of predictive analytics are used in different settings and contexts.

Predictive Models

The most common model type is the predictive model. Predictive models aim to make a prediction by comparing something in a group to another thing in a different group ^[7]. By comparing attributes and features of a specific thing, you can make predictions about how another thing will behave with a similar set of circumstances. The attributes and features form the variables of the model. Prediction models are built by feeding the model training data which will tell be used for comparison against a "out-of-training sample." The training data would contain lists of verifiable data (such as pictures of spiders with the gender tagged) and then would compare those to the "out-of-training sample" pictures of spiders without the gender tagged. The model would then make predictions about the spider's gender based on the training data.

Descriptive Models

Descriptive models differ from predictive models in that the focus is on groups rather than the individual. Predictive models seek to make a prediction about the behavior of one thing, whereas descriptive models seek to categorize individual things into groups ^[8]. Descriptive models generalize relationships between a large group of things, generally to categorize and organize things by a variable. Descriptive models are typically used to make inferences about how groups would behave and their preferences.

Decision Models

Decision models are complex and are used for predicting the results of a decision ^[9]. Decision models take in data from multiple sources, including predictive and descriptive models, and use that to predict outcomes. Predictive models can predict individual outcomes, such as an individual's credit risk, but cannot predict what would happen to the individual if they were denied or approved for credit. Decision models make these complex predictions.

Uses

Predictive analytics have multiple applications and are used in a variety of contexts.

Healthcare

Graphic demonstrating how patient data gets fed into a model to generate predictive insights on patients. ^[10]

Healthcare analytics refers to systematic use of health data and related business insights developed through applying analytics to drive fact-based decision making for planning, management, measurement, and learning in healthcare. ^[11] Predictive analytics involves using similar methods to generate data predictions and other techniques for assessing predictive power. ^[12] In the context of healthcare, it can be used to identify high-risk patients and provide treatment, reducing unnecessary hospitalizations or readmissions. Reserachers Harris et. al. developed an analytical model to predict future patient behavior based on past behavior. ^[13] This model provides an accurate prediction of no-show patients and assists clinics in developing operational mitigation strategies such as overbooking appointment slots and managing patients predicted as “no-shows”. ^[14] Models such as these can be used for clinical planning and scheduling decisions to improve patient service at hospitals with optimal and unique solutions. ^[15]

In healthcare, predictive analytics have been used to study Parkinson's disease.^[16] Additionally, some support its use for creating models that predict people who could have a greater chance of developing chronic diseases, thus helping them identify these illnesses earlier on, saving them time and money.^[17]

Human Resources

In the human resources field, predictive analytics and modeling can be used for things like forecasting openings within companies and anticipating which employees are a liability.^[18] The use of predictive analytics in human resources has recently been increasing in popularity. ^[19] Companies use data and analytics to design, evaluate, and implement new management policies; this also means that the traditional methods of using experience, intuition, and guesswork to guid human resources strategy is falling to the wayside. ^[20] Applying HR analytics in a firm can be a one-time effort that gradually becomes a newly overhauled approach to organizational management and it is not uncommon that these inspire more broad-reaching change to a corporation. ^[21] Analytics must be rooted in understanding the data to be used and the context under which the data were collected to identify any potential biases. ^[22]

A study conducted in 2019 of 4,800 individuals across different companies in different industries and determined that roughly one-quarter to one-third of all companies use predictive analytics in human resources. ^[23] From this study, the industry that uses predictive analytics the most in human resources is financial services, with 32% of companies applying analytics. ^[24] The Technology (software), Oil and Energy, and Healthcare and Pharmaceuticals industries all had over 25% of companies applying analytics to human resources. ^[25]

Law Enforcement

In law enforcement, predictive modeling techniques have been used and referred to as PredPol (derived from the term Predictive Policing), as in the Santa Cruz California Police Department. Officers at this department state that it is used as a supplementary tool rather than a replacement for their normal rotations. Additionally, this PredPol system predicts solely based on crimes reported and not demographic or identifying information of individuals involved in the crimes in an attempt to reduce demographically based biases.^[26]

Ethical Challenges

Bias and Discrimination

Ever since the rise of the computer gaming industry brought back the resurgence of neural networks, experts have argued that deep learning is a highly effective way to train an artificial intelligence system.^[27] Designed to mimic the way a human brain thinks and makes decisions, a network of thousands or millions of individual processing nodes are connected together in a neural net, which enables an algorithm to train itself to perform a task given a prepared training data set.^[27] However, according to Barocas and Selbst from Cornell University and UCLA respectively, “an algorithm is only as good as the data it works with."^[28] Zarsky, a professor and vice dean at Haifa University, argues that algorithms trained on biased data sets will not only inherit pre-existing biases from the aforementioned data set but also generate novel patterns of unfair bias and discrimination and reinforce these patterns in their decision-making processes.^[29] Some surmise an algorithm may interpret inequalities in historical data as sensible patterns, which support existing societal biases. ^[28]. Detecting and addressing unfair bias and discrimination in algorithms for predictive analytics is particularly difficult as it occurs due to unintended consequences from using the algorithm, not the purposeful actions of an ill-intentioned programmer.^[28]

Transparency

Some assert that transparency as an ethical issue is in opposition to other ethical interests such as privacy.^[30] Transparency in algorithms means that the algorithm should not only have its details accessible but also comprehensible to humans analyzing them; accessible information that is not decipherable is no longer useful.^[31] Many modern sophisticated artificial intelligence systems are trained via deep learning that is comprised of extensive neural networks reaching up to fifty layers.^[27] As each layer adds complexity, Sloan and Warner assert that the human comprehensibility of these networks is affected and thus the transparency.^[32]

Predictive Privacy

The term “predictive privacy” refers to the ethical challenges posed by the ability of algorithms to predict sensitive information about an individual using information derived from data sets of other individuals.^[5] In 2019, the Electronic Privacy Information Center (EPIC) raised this ethical concern in their official complaint to the Federal Trade Commission (FTC) against HireVue, a recruiting-technology company, stating that “the company’s use of unproven artificial-intelligence systems that scan people’s faces and voices [constitutes] a wide-scale threat to American workers."^[3] Mühlhoff’s definition of a violation of predictive privacy is “if sensitive information about [a] person or group is predicted against their will or without their knowledge on the basis of data of many other individuals, provided that these predictions lead to decisions that affect anyone’s...freedom."^[5] Some note that predictive privacy can be violated regardless of the prediction’s accuracy, especially when systems for data collection and processing are designed such that subjects cannot provide meaningful or informed consent.^[33]

References

1. ↑ Nyce, Charles. "Predictive Analytics White Paper." The Digital Insurer, American Institute for CPCU, 2007, www.the-digital-insurer.com/wp-content/uploads/2013/12/78-Predictive-Modeling-White-Paper.pdf.
2. ↑ Cohen, I. G., et al. "The Legal And Ethical Concerns That Arise From Using Complex Predictive Analytics In Health Care." Health Affairs, vol. 33, no. 7, 2014, pp. 1139-47, doi:10.1377/hlthaff.2014.0048.
3. ↑ ^{3.0 3.1} Harwell, Drew. "Rights group files federal complaint against AI-hiring firm HireVue, citing ‘unfair and deceptive’ practices." Washington Post, 6 November 2019, www.washingtonpost.com/technology/2019/11/06/prominent-rights-group-files-federal-complaint-against-ai-hiring-firm-hirevue-citing-unfair-deceptive-practices.
4. ↑ Perry, Walter, et al. "Predictive Policing: The Role Of Crime Forecasting In Law Enforcement Operations." RAND Corporation, 2013, doi:10.7249/rr233.
5. ↑ ^{5.0 5.1 5.2} Mühlhoff, Rainer. "Predictive Privacy: Towards An Applied Ethics Of Data Analytics." SSRN, 2020, doi:10.2139/ssrn.3724185.
6. ↑ Bachar, Daniel. "Descriptive, Predictive and Prescriptive Analytics Explained ." Logility, https://www.logility.com/blog/descriptive-predictive-and-prescriptive-analytics-explained/. Accessed 9 April 2021.
7. ↑ Carew, Joseph. "What is Predictive Modeling." TechTarget, December 2020, https://searchenterpriseai.techtarget.com/definition/predictive-modeling. Accessed 9 April 2021.
8. ↑ Artale A. "A Descriptive Model." Rings in Auctions., 1997, https://doi.org/10.1007/978-3-642-59158-7_3. Accessed 9 April 2021.
9. ↑ Kuntz K, Sainfort F, Butler M, et al. "Overview of Decision Models Used in Research." NCBI, February 2013, https://www.ncbi.nlm.nih.gov/books/NBK127474/. Accessed 9 April 2021.
10. ↑ Lynn, John. “Using NLP with Machine Learning for Predictive Analytics in Healthcare”. Healthcare IT Today. December 12, 2016
11. ↑ Kankanhalli, Atreyi, et al. "Big data and analytics in healthcare: introduction to the special section." Information Systems Frontiers 18.2 (2016): 233-235.
12. ↑
13. ↑ Harris, Shannon L., Jerrold H. May, and Luis G. Vargas. "Predictive analytics model for healthcare planning and scheduling." European Journal of Operational Research 253.1 (2016): 121-131.
14. ↑
15. ↑
16. ↑ Dinov, Ivo D., et al. "Predictive Big Data Analytics: A Study of Parkinson’s Disease Using Large, Complex, Heterogeneous, Incongruent, Multi-Source and Incomplete Observations." PLoS One, vol. 11, no. 8, 5 August 2016, doi:10.1371/journal.pone.0157077.
17. ↑ "Predictive analytics in healthcare." Foresee Medical, www.foreseemed.com/predictive-analytics-in-healthcare. Accessed 28 March 2021.
18. ↑ Mishra, Sujeet N., et al. "Human Resource Predictive Analytics (HRPA) for HR Management in Organizations." International Journal of Scientific & Technology Research, vol. 5, no. 5, May 2016, www.ijstr.org/final-print/may2016/Human-Resource-Predictive-Analytics-hrpa-For-Hr-Management-In-Organizations.pdf.
19. ↑ King, Kylie Goodell. "Data analytics in human resources: A case study and critical review." Human Resource Development Review 15.4 (2016): 487-495.
20. ↑
21. ↑
22. ↑
23. ↑ Noack, Brent. "Big data analytics in human resource management: Automated decision-making processes, predictive hiring algorithms, and cutting-edge workplace surveillance technologies." Psychosociological Issues in Human Resource Management 7.2 (2019): 37-42.
24. ↑
25. ↑
26. ↑ Eidam, Eyragon. "The Role of Data Analytics in Predictive Policing." Government Technology, September 2016, www.govtech.com/data/Role-of-Data-Analytics-in-Predictive-Policing.html. Accessed 28 March 2021.
27. ↑ ^{27.0 27.1 27.2} Hardesty, Larry. "Explained: Neural Networks." MIT News, 2021, news.mit.edu/2017/explained-neural-networks-deep-learning-0414.
28. ↑ ^{28.0 28.1 28.2} Barocas, Solon, and Andrew D. Selbst. "Big Data's Disparate Impact." SSRN, 2016, doi:10.2139/ssrn.2477899.
29. ↑ Zarsky, Tal Z. "An Analytic Challenge: Discrimination Theory in the Age of Predictive Analytics." I/S: A Journal of Law and Policy, vol. 14.1, 2017, pp. 12-35, kb.osu.edu/bitstream/handle/1811/86702/1/ISJLP_V14N1_011.pdf.
30. ↑ Canca, Cansu. "Anonymity in the Time of a Pandemic: Privacy vs. Transparency." Bill of Health, Harvard Law, blog.petrieflom.law.harvard.edu/2020/03/30/anonymity-in-the-time-of-a-pandemic-privacy-vs-transparency. Accessed 27 March 2021.
31. ↑ Mittelstadt, Brent D., et al. "The Ethics Of Algorithms: Mapping The Debate." Big Data & Society, vol. 3, no. 2, 2016, pp. 1-21, SAGE Publications, doi:10.1177/2053951716679679.
32. ↑ Sloan, Robert H., Richard Warner. "When Is an Algorithm Transparent?: Predictive Analytics, Privacy, and Public Policy." IEEE: Security & Privacy, SSRN, 2017, dx.doi.org/10.2139/ssrn.3051588.
33. ↑ Schermer, Bart W. "The Limits Of Privacy In Automated Profiling And Data Mining." Computer Law & Security Review, vol. 27, no. 1, 2011, pp. 45-52, doi:10.1016/j.clsr.2010.11.009.