Security Issues In Predictive Analytics Essay


Discuss about the Security and Privacy Issues in Predictive Analytics.


Technology Description

Predictive Analytics is a branch within advanced analytics used to make predictions about future events that are still unknown. Predictive analytics makes use of several techniques including statistics, data mining, machine learning, modeling, as well as AI (artificial intelligence) to analyze present data and make predictions concerning the future. Predictive analytics makes use of various predictive modeling, data mining, and analysis techniques to bring together information technology, management, and business modeling processes to make future predictions. The technology works on the basis of transactional and historical data that are used for purposes such as identifying future risks or identifying future opportunities. Models used in predictive analysis capture relationships among several factors so as to assess risks using a specific set of conditions to assign weights or scores (Waller & Fawcett 2013).

Business users, utilizing text analytics, data mining, as well as statistics crate predictive intelligence through the uncovering of relationships and patterns in both unstructured and structured data. Structured data, such as gender, age, marital status, and sales, can be readily used for predictive analysis. Unstructured data, such as social media content, text data in call center information/ notes, sentiment such as those in social media or open text that must be extracted are also used for predictive analytics in the process of model building (Umachandran & Ferdinand-James 2017). Using predictive analytics, organizations become forward looking, proactive, and anticipate behaviors and outcomes based on data and not on assumptions or just hunches. The result is better decision making and planning, which among other things confer competitive advantages to the organization (Rickman & Cosenza 2007).

Technology Solution Assessment

While predictive analytics greatly helps businesses such as to plan for the future for example in projecting future demand, consumption patterns among different age groups, and possible new products, it comes with risks, especially with regard to consumer data and information (Crawford & Schultz 2014). To achieve predictive analytics, organizations inevitably must make use of consumer data, such as customer gender, age, and sentiments in social media in which case both structured and unstructured data is used for predictive analytics. Big data is, and continues to carry enormous potential for revolutionizing people's lives through predictive power for example, it is possible to accurately predict weather to 95% accuracy 48 hours before the time (Murphy 2015). The sheer scales of people that have been involved in security incidents of big data, the stakes are even higher. For instance, the breach of eBay data in 2014 resulted in a breach to personal information for 145 million people whose e-mail and home addresses as well as birth dates were exposed/ breached (Finkle, Chatterjee and Maan, 2014). Because of the sheer size, the different aspects of the data as well as its diverse sources, it is also a daunting task to protect this amount of data. Vulnerability to unauthorized access is multiplied because of the broad and distributed range of access (Terzi, Terzi & Sagiroglu 2015).

Privacy Issues

Breaches to privacy that cause embarrassments and other risks: Actions that organizations take in using big data for analytics can easily breach the privacy of the users’ data and result in embarrassments and law suits, as well as loss of jobs. Some retailers, for instance, use big data on customers, making use of such details as pregnant customers due dates, or even monthly menstrual cycles for women, or the color of lingerie most bought by a customer (Crawford & Schultz 2014).

Data masking being defeated so as to reveal personal information: If data masking is used inappropriately, predictive analytics of big data can reveal persons whose data was masked. The newness of predictive analytics of big data means that organizations still remain aware of such risks, greatly exposing individual private information to unwanted audiences, such as hackers

Individuals have no (or there are very few) legal protections for them. While authorities and regulators have expressed risks to privacy due to predictive analytics of bid data, legal requirements for protecting privacy when big data analytics are being undertaken are not yet existent, or remain unclear and opaque (Paulson & Scruth 2017).

Risk of unethical decisions from predictive analytics of big data

Predictive analytics of big data can be employed to influence behavior which is unethical and this happens when organizations use big data analytics to make decisions that do not take cognizance of the value of human life/ health. The potential for revealing people’s personal information since it is not illegal through this revelation can damage the lives/ health of the concerned persons poses another privacy hazard (Kshetri 2014).

Discriminatory tendencies: Predictive analytics of big data can be used to provide promotions, develop courses among other uses; the results can back fire if there is no objectivity. Big data can make discrimination more prevalent and pervasive, for instance, in human resource planning. A financial institution cannot determine the sexual orientation or race of a loan applicant since this is illegal in the first place; however, using big data and predictive analytics, the race, gender, age, financial situation, even address of the prospective applicant can be mined through big data analytics by mining such information from the Internet of Things or from on-line platforms such as social media (Loehr 2017). With big data analytics, a loan request can be turned down based simply on discriminatory decision making or algorithms that are inherently discriminatory.

Security concerns

Predictive analytics not always accurate: Despite the use of predictive analytics of big data, some issues may not be unearthed since the technology is not fully accurate. The data files employed in predictive analysis can contain inaccurate information and data on individuals or the algorithms can be flawed. Predictive analytics are only as good as the computations used for generating results. The risk of inaccuracies increases proportionately to the addition of data files to existing datasets along with the use of complex models for data analysis. Financial firms such as VISA rely a great deal on predictive big data analytics, for among other things, detecting security breaches and fraud, but there are problems with the models, fraud will still happen (Armerding 2017).

The large amounts of data organizations collect about individuals and store distributive, for instance Amazon, creates enhanced risks to the security of the data. The data can be stolen or hacked and be used for further malicious attacks. Predictive analytics is being used, and offers huge promises for organizations in predicting employee behavior; the risk of flight can be predicted early and appropriate measures taken. However, a new element is introduced with such applications; speculative data on employees. Beyond the standard financial and personal information on employees, predictive analysis crates a new problem of future behavior estimation, speaking to the mind, heart, and intentions of this employee. The questions that arise are ethical and practical; what is the predictive analysis on flight risks are wrong? The prediction of behavior can result in targeted responses by HR which borders in mind control and behavioral modification, a grave ethical issue. If the analytics are wrong, an employee can be wrongly labeled as being disloyal and have their reputation blemished, yet in reality, they may just have different behavior and this can result in unfair actions from management. Apart from possibly giving wrong conclusions, it is a pervasive privacy invasion; information that can be used maliciously even by insiders in the organization.

Organizational Change Assessment

A survey shows that several big organizations still believe that big data has several huge untapped opportunities for the future. Big data is being viewed from the context of different datasets integration to uncover or drive specific insights. A third of the respondents believe that analytics will be an integral driver or organizational change and transformation within organizations, while also forming a significant part of day to day activities in the running of organizations (Klein 2014). Predictive analytics of big data or any other data will have a significant impact in organizational processes, including in product development. For instance, a software or application developer will change how they develop software and the speed with which this software is developed and delivered based on predictive analytics. Predictive analytics is presently used widely in the financial services sector, for instance to predict credit risk of a client, by insurance firms for predicting losses, by law enforcement agencies to predict the nature and kinds of criminal acts, and by organizations to predict employee behavior and flight risk, as HP already does (Siegel 2013) . Predictive analytics can be expanded to prescriptive analytics so an individual can know what is likely to cause them problems in future and what they can do to overcome such problems. Using predictive analytics to determine future customer behavior, product and market trends, or employee behavior, organizations will (and already are) gaining strategic competitive advantages in their markets and beating the competition not yet using predictive analytics comprehensively. Already, predictive analytics has been used to predict winners of elections; Nate Silver predicted, back in 2012, the winner in all the fifty DC states and has also accurately predicted nine presidential elections before 2012, even before predictive analytics of big data came into the mainstream. For any commercial organization, there is information related to the customer, such as customer satisfaction, expectations, social media activity, referrals, and service levels; there is also financial information including profitability, revenue, margins, and competitor performance. There is also operational information such as cycle time, productivity, errors, waste, among others, while information related to the workforce such as skill acquisition and effectiveness of training also exist.

Predictive analytics can be applied to help solve some of the most challenging aspects of an organization; one prominent area where predictive analytics can be beneficial is in procurement (Galvan 2015). The need for efficiency and getting the best value through procurement is a pressing one for both private and public organizations. Traditionally, legacy systems have been used to manage procurement; however, the ERP systems provide few if any, insights into business processes and usually are the cause of-instead of the solution to-bottlenecks in the procurement chain (order to delivery). This happens because traditional processes of procurement do not have real time data analysis capacity that provides valuable insights to help with decision making. The traditional procurement methods rely heavily on human intelligence, which has its challenges and imperfections to drive performance, cost savings, and efficiency. In contrast, predictive analytics gathers large volumes of data in real time on delivery networks, supply chains, customer sales, and billing and uses powerful computer algorithms to mine trends, insights, and other forms of intelligence. Predictive analytics manages to undertake these analytics continuously and in real time and this implies that firms can apply them immediately to reduce costs, improve performance, and attain higher levels of efficiency in their procurement processes. Using predictive analytics, organizations will gain better intelligence into supply chains; these chains provide tons of useful data from several tracking systems, audits, and inspections. The generated data is difficult to analyze by hand because of the high volume but through the use of computer analytics, holistic and meaningful conclusions can be derived. With this information, firms can better plan and optimize procurement scenarios, undertake more accurate demand forecasting, and collaborate better with suppliers, consequently enabling better planning to be integrated into the whole organization.

Any business organization can be transformed through the use of predictive analytics because of knowing likely outcomes, trends, scenarios with regard to the success of the organization. Predictive analytics can significantly change and transform any organization across any industry. In the health care sector, predictive analytics is transforming health care by enabling organizations (hospitals) to spot trends in diverse areas, from staffing to needs for readmission. Health organizations are hiring specialist data analysts to transform health care by applying certain algorithms to collected data on health care and generating useful insights from such data, according to Hede (2016). The health care industry can especially benefit from predictive analytics in a great way because most organizations in this industry long ago adopted the concept of EHR (electronic health records) and so have large treasure troves of highly useful data. In the crucial cyber security sector, players are realizing the power of harnessing predictive analytics by evaluating aspects like Internet chatter using specific algorithms to identify patterns and analyzing past attacks and incidents to identify and root out/ prepare better for possible attacks (Amjad 2016). The construction industry too is not left behind; being organizations that undertake some of the biggest and most complex construction project in the world; number crunching comes naturally to these organizations. However, nearly 35% of wastes in the construction industry are attributed to material waste; using big data analytics, construction companies can, and have been able to significantly lower costs through the use of predictive analytics in project management (Marr 2016). Research shows that organizational transformation wrought about by big data and predictive analytics is field executives are taking more seriously and planning for the inevitable transformation of their organizations due to this phenomenon (Klein 2014).


Predictive analytics is a form of advanced analytics used to make predictions about events that are still unknown. Techniques used in predictive analytics of big data include statistics, data mining, machine learning, modeling, as well as AI (artificial intelligence). It’s therefore beneficial to organizations for future planning, responding to market changes, projecting future demand and consumption patterns among other myriad benefits. Predictive analytics brings the risks of privacy, data security, and even discriminatory tendencies when used for predicting employee loyalty, for example. Predictive analytics is forcing companies to change their organizational culture; rather than the firms changing their culture because of the huge benefits from predictive analytics. Organizations across industries, ranging from health care, security, cyber security, manufacturing, retail, financial and construction are being transformed by predictive analytics. This paper concludes therefore, that predictive analytics is new and high disruptive phenomenon that has numerous benefits for industries, from planning operations to responding to future market threats. However, it has some issues with regard to security and privacy of personal data and information, which is what, is used for analysis and making predictions. Predictive analytics is however forcing organizations to change their processes just to stay competitive.


Armerding, T. (2017). The 5 worst big data privacy risks (and how to guard against them). [online] CSO Online. Available at: [Accessed 1 Oct. 2017].

Crawford, K., & Schultz, J. (2014). Big Data and Due Process: Toward a Framework to Redress Predictive Privacy Harms. Boston College Law Review. Boston College Law School.

Demurjian, S. A., & Moussa, M. A. (2017). Differential Privacy Approach for Big Data Privacy in Healthcare.

Finkle, J., Chatterjee, S. and Maan, L. (2014). EBay asks 145 million users to change passwords after cyber attack. [online] Reuters. Available at: [Accessed 1 Oct. 2017].

Galvan, J. (2015). Predictive analytics can transform any type of organization. [online] IBM Big Data & Analytics Hub. Available at: [Accessed 1 Oct. 2017].

Kshetri, Nir B. (2014). Big Data's Impact on Privacy, Security and Consumer Welfare.

Klein, L. (2014). Future of Analytics: Big Data Integration, Transforming Organizations and Processes, Providing Speed and Foresight - Predictive Analytics Times - machine learning & data science news. [online] Predictive Analytics Times. Available at: [Accessed 1 Oct. 2017].

Loehr, A. (2017). Big Data for HR: Can Predictive Analyltics Decrease Discrimination?. [online] Available at: [Accessed 1 Oct. 2017].

Marr, B. (2016). How Big Data And Analytics Are Transforming The Construction Industry. [online] Available at: [Accessed 1 Oct. 2017].

Murphy, M. (2015). IBM is going to change how we forecast the weather with Watson. [online] Quartz. Available at: [Accessed 1 Oct. 2017].

Paulson SS, & Scruth E. (2017). Legal and Ethical Concerns of Big Data: Predictive Analytics. Clinical Nurse Specialist CNS. 31, 237-239.

Rickman, T. A., & Cosenza, R. M. (2007). The changing digital dynamics of multichannel marketing: The feasibility of the weblog: text mining approach for fast fashion trending. Journal Of Fashion Marketing And Management. 11, 604-621.

Siegel, E. (2013). Predictive Analytics: The privacy pickle - Hewlett-Packard’s prediction of employee behavior - Analytics Magazine. [online] Analytics Magazine. Available at: [Accessed 1 Oct. 2017].

Terzi, D. S., Terzi, R., & Sagiroglu, S. (2015). A survey on security and privacy issues in big data. 202-207.

Umachandran, K., & Ferdinand-James, D. S. (2017). Affordances of Data Science in Agriculture, Manufacturing, and Education. InfoSci-Books

Waller M.A., & Fawcett S.E. (2013). Data science, predictive analytics, and big data: A revolution that will transform supply chain design and management. Journal of Business Logistics. 34, 77-84.

How to cite this essay: