Machine Learning Times
Machine Learning Times
EXCLUSIVE HIGHLIGHTS
SHARE THIS:

7 years ago
Data Repurposing: The Underpinning of Predictive Analytics?

 

C-suite executives are at once enamored and afraid of big data. The notion of information created moment-to-moment and with every action of a company hints at massive analytics opportunity, but it can seem simultaneously overwhelming. Increasingly, IT professionals turn to robust analytics tools to handle this data and to derive prescriptive solutions for existing problems, but that is just the tip of the iceberg. Data repurposing — using previously collected data in a new way to extract fresh insights — is gaining ground as a way to predict outcomes rather than simply to prescribe remedies.

The Unconventional Attraction

According to a November 20 article in Forbes, big data has moved beyond buzzword status to become a full-fledged movement. While entry-level access to big data benefits means using collected information to create high-level metrics and to improve business intelligence, co-founder Mayank Bawa of analytics firm Teradata Aster said that the big data movement is really “about how organizations leverage different kinds of information for different types of business purposes to unlock value that was previously unattainable or unknowable.” The right analytics solution lets companies find value in what Bawa terms “unconventional” data points, such as social networking data, text messaging or collaborative documents.

Companies cannot ignore what they have already collected; data repurposing gives previously stored data new life when better questions are asked or unconventional data points are considered as part of a larger whole. Ideally, this insight provides the foundation for what is known as predictive analytics, or educated expectations based on current data for any aspect of a company’s future.

The HR Connection

One area in which this kind of analytics has seen growth is human resources (HR). As a recent InformationWeek article notes, while it is difficult to predict what an individual employee will do over a given period, it is becoming possible to predict group behavior, such as that of large call centers. By building into HR solutions the types of analytics tools that collect not only core employee data such as late arrivals, absences and overtime hours worked but also social data from both public profiles and internal networks, midsize companies can predict how many employees will leave in a month or a year and how many will stay the course. Privacy is the limiting factor, and attempting to anticipate the future actions of an individual could spark backlash. As a broad-spectrum set, however, this kind of data collection and repurposing helps to generate a solid predictive picture.

For midsize IT professionals and administrators, this wider interest in analytics means increased expectations both for existing data to generate new insights and new data to be effectively integrated even though it is largely unstructured. To accomplish C-suite and HR aims, IT departments should advocate for full-featured analytics tools or the support to build such tools in house. Gathering approval — and getting a budget approved — means focusing on the large, virtually untapped resource of previously collected data. With the right investment and intelligent data repurposing, midsize businesses can anticipate rather than react to market and employee forces, and that is always good news for ROI.

By: Doug Bonderud, Freelance Writer
Originally published at midsizeinsider
This post was written as part of the IBM for Midsize Business program, which provides midsize businesses with the tools, expertise and solutions they need to become engines of a smarter planet.

Leave a Reply