Avoid Non-compliance Charges as You Mine Your Data | DCR Workforce Blog

Avoid Non-compliance Charges as You Mine Your Data

What is data mining and what exactly do organizations hope to gain from mining their data? Is data mining about efficient use of technology, or making business decisions which add a lot of value? Collecting data is only the first step. It is of no use to anyone if they collect the data and dump it in an obscure corner and never look at it. Data could have been buried for so long that no one currently making decisions in the organization has any clue to its existence. On the other hand, how many companies intuitively know how to mine their raw data and derive insights which automatically make sense? How many expend a lot of energy on doing so and get disheartened and decide to drop it as a waste of time, in a huff, not to bother with it anymore? If they do choose to press ahead with their efforts, will they definitely reap the benefits touted? Is there any legal necessity which binds their actions as they engage in data mining? So many questions and very few answers! Let us see if we can throw any light on at least some of them.

Data mining provides insights by identifying anomalies, patterns and correlations between different (very large) data sets and helps to predict possible outcomes through the correlation. Done right, it could provide insights which reduce costs, improve revenues, control attrition, reduce risks, enable business decisions across multiple locations and improve regulatory compliance.

Some critical steps in data mining:

Before initiating the necessary efforts to make your business data-driven, do ensure that some critical initiative assessment is undertaken to decide what your needs will be:

  • How much data does your business generate and how much of it is valuable and come in useful when setting the strategy for future business activities using data-driven insights?
  • Do you collect the data and do have enough data collected for purposes of analysis? Did you already search through some of it, looking for insights? Do you feel you or your teams are wasting a great deal of their time (say 15% to 20%) on the task of searching for critical information because your data is unstructured, chaotic and just a lot of repetitive noise?
  • Do you have no data collected and taking decisions in a vacuum-filled business environment where no learning is derived from any of the old successes or failures?
  • Have you included your engineering and business teams in the effort being made by your IT team to collect, structure and analyze the data?
  • Have you identified the datasets which are critical to the analysis being planned? Estimated their size? Configured enough data warehousing and storage for them with the necessary access points? You may also need to decide whether to create the storage locally or remotely and decide if you need to set up appropriated access credentials and firewalls to keep out anyone without the required authorization.

Now is the time to strike a note of caution! These preliminary strategy meetings must also recognize the benefits and risks which come with data mining and data driven decision making. It all depends on where you plan to use big data. Is it to hire and manage your workforces, or is it in marketing, or in fraud prevention? Are you sure of the quality of data, in terms of its representativeness, accuracy, robustness, reliability and completeness?

  • With human resource planning using big data, there is a pressing need to ensure that it does not become tool for inclusion or exclusion, as the case may be! Setting the proxies for performance or success can be linked to attributes like work ethic for existing workers, but assuming much based on a resume‚Äôs contents could prove risky.
  • Eliminate the possibility of having any personal biases creeping in, into the data-linked decisions; which could result in discrimination and unequal treatment.
  • Exercise adequate care to identify, address and prevent the possible sources of discrimination?
  • Be aware of the tenets of these laws which come into consideration, when using big data, to eliminate possible discrimination: Federal Trade Commission Act (FTCA), Federal Credit Reporting Act (FCRA), Equal Opportunity Laws, Title VII of the Civil Rights Act, Americans with Disabilities Act, Age Discrimination in Employment Act (ADEA), Fair Housing Act, Genetic Information Nondiscrimination Act (GINA).

At all times, employers need to ensure that their data set is representative and that their data model eliminates possible biases as well as ethical and fairness concerns. They need to also stay alert to the accuracy (or lack thereof) of any predictions made by their big data mining efforts and revise them continually until they prove reliable.


Disclaimer:
The content on this blog is for informational purposes only and cannot be construed as specific legal advice or as a substitute for competent legal advice. They reflect the opinions of DCR Workforce and may not reflect the opinions of any individual attorney. Do contact an attorney for advice specific to your issue or problem.
Lalita is a people/project manager with extensive experience in operations, HCM and training and development across industries like banking, education, business consulting, BPO and information technology. She believes in a dynamic approach to life and learning as change is the only constant.