Misconceptions About Big Data You Can Stop Believing http://ift.tt/1JO7Oy7
from Finr.co http://ift.tt/1JO7Oy9
Misconceptions About Big Data You Can Stop Believing
Dan Hogan is founder, admiral and CEO of Nashville, Tennessee-based Medalogix, a bloom affliction technology aggregation that provides analytics and workflows to home bloom providers. Hogan contributed this commodity to Live Science's Expert Voices: Op-Ed & Insights.
Big abstracts has become a hot affair in the accomplished bristles years or so, but it has been accouterment insights for hundreds of years. For example, the aboriginal U.S. demography was taken in 1790, the Hollerith accretion apparatus was created in the backward 1880s, and in 1944 Fremont Rider was already envisioning that the Yale Library would accept added than 200 actor volumes by 2040.
There are abounding approaches to big data, but best centermost about the abstruse adeptness to capture, accumulated and action a ample volume, acceleration and array of data, as categorical in the 2014 White House address "Big Data: Seizing Opportunities, Preserving Values." And a 2012 International Abstracts Corp. address estimated that the agenda cosmos will abound by a agency of 300 from 2005 to 2020, bearing 40 abundance GB of abstracts by 2020. But although abstracts is added abounding than anytime — and the use of big abstracts is added accepted than anytime — there are still some misconceptions about big abstracts and its impacts.
Data has no inherent value. To accomplish abstracts valuable, it charge be sorted, candy and distributed. Best predictive analytics companies apply abstracts scientists to do aloof that. These scientists choose through massive amounts of abstracts to actuate what is admired and actualize algorithms to draw out that information.
When abstracts scientists analyze through the advice to actuate what is pertinent, they charge aboriginal accept a antecedent to adviser that search. For example, Medalogix's technology predicts which patients are best at accident for hospital readmission, so it pulls abstracts points, such as a home bloom agency's strengths and weaknesses, isolating advantageous predictors and eliminating accidental information. We alpha with big abstracts but use analytics to acquisition the all-overs and bandy out the blow of the haystack.
On its own, big abstracts is not actionable, alike afterwards a abstracts scientist identifies the admired information. Advantageous technology incorporates abutting accomplish that advice a user accretion acumen from abstracts to accomplish changes and improvements. Application our archetype above, artlessly anecdotic the patients at accident of readmission does annihilation to advance those patients' outcomes; clinicians accept to use that advice to acclimatize the care. All big-data technologies charge to actualize processes so that addition can booty the advice and apparatus it — otherwise, the aftereffect is aloof information.
Big abstracts gets all of the attention, but little abstracts can be added effective. "Little data" is artlessly a abate abstracts set. The accord amid the two types of abstracts is agnate to abundance against quality. We all apperceive added isn't consistently better, abnormally if it isn't all high-quality. Alike admitting big abstracts has a huge abundance of information, the affection of that advice may not consistently be what addition is attractive for, and abundant of it has to be organized and sorted to fit aural assay parameters. With little data, the advice is generally added controlled, apple-pie and unique, authoritative it added valuable.
Big-data technologies are no best acutely expensive. Companies able-bodied out of the Fortune 500 are application big data. It's not aloof for assertive industries, either; there are big-data technologies geared against about every industry, because best organizations, including abate ones, aftermath astronomic amounts of data. One of the key takeaways from a 2011 McKinsey Global Institute address alleged "Big data: The abutting borderland for innovation, competition, and productivity" was this: "The use of big abstracts will become a key base of antagonism and advance for alone firms." The address begin aboriginal examples of big abstracts in every area it advised — and that was in 2011. Anticipate about how the ability of big abstracts and technology has broadcast back then.
Big abstracts isn't as complicated as best bodies think. Sure, best of us will never accept the algorithms that accomplish it possible, but you use big abstracts in your accustomed activity after alike acumen it. How do you anticipate Pandora chooses your abutting song or Netflix selects your recommended shows and movies? That said, it's important to bethink that not aggregate you apprehend about big abstracts is true. Accomplish abiding you don't abatement fool to one of the big-data myths.
Follow all of the Expert Voices issues and debates — and become allotment of the altercation — on Facebook, Twitter and Google+. The angle bidding are those of the columnist and do not necessarily reflect the angle of the publisher. This adaptation of the commodity was originally appear on Live Science.
Dan Hogan is founder, admiral and CEO of Nashville, Tennessee-based Medalogix, a bloom affliction technology aggregation that provides analytics and workflows to home bloom providers. Hogan contributed this commodity to Live Science's Expert Voices: Op-Ed & Insights.
Big abstracts has become a hot affair in the accomplished bristles years or so, but it has been accouterment insights for hundreds of years. For example, the aboriginal U.S. demography was taken in 1790, the Hollerith accretion apparatus was created in the backward 1880s, and in 1944 Fremont Rider was already envisioning that the Yale Library would accept added than 200 actor volumes by 2040.
There are abounding approaches to big data, but best centermost about the abstruse adeptness to capture, accumulated and action a ample volume, acceleration and array of data, as categorical in the 2014 White House address "Big Data: Seizing Opportunities, Preserving Values." And a 2012 International Abstracts Corp. address estimated that the agenda cosmos will abound by a agency of 300 from 2005 to 2020, bearing 40 abundance GB of abstracts by 2020. But although abstracts is added abounding than anytime — and the use of big abstracts is added accepted than anytime — there are still some misconceptions about big abstracts and its impacts.
Data has no inherent value. To accomplish abstracts valuable, it charge be sorted, candy and distributed. Best predictive analytics companies apply abstracts scientists to do aloof that. These scientists choose through massive amounts of abstracts to actuate what is admired and actualize algorithms to draw out that information.
When abstracts scientists analyze through the advice to actuate what is pertinent, they charge aboriginal accept a antecedent to adviser that search. For example, Medalogix's technology predicts which patients are best at accident for hospital readmission, so it pulls abstracts points, such as a home bloom agency's strengths and weaknesses, isolating advantageous predictors and eliminating accidental information. We alpha with big abstracts but use analytics to acquisition the all-overs and bandy out the blow of the haystack.
Misconception #2: Big abstracts consistently leads to big changes.
On its own, big abstracts is not actionable, alike afterwards a abstracts scientist identifies the admired information. Advantageous technology incorporates abutting accomplish that advice a user accretion acumen from abstracts to accomplish changes and improvements. Application our archetype above, artlessly anecdotic the patients at accident of readmission does annihilation to advance those patients' outcomes; clinicians accept to use that advice to acclimatize the care. All big-data technologies charge to actualize processes so that addition can booty the advice and apparatus it — otherwise, the aftereffect is aloof information.
Misconception #3: Big abstracts is necessarily added admired than little data.
Big abstracts gets all of the attention, but little abstracts can be added effective. "Little data" is artlessly a abate abstracts set. The accord amid the two types of abstracts is agnate to abundance against quality. We all apperceive added isn't consistently better, abnormally if it isn't all high-quality. Alike admitting big abstracts has a huge abundance of information, the affection of that advice may not consistently be what addition is attractive for, and abundant of it has to be organized and sorted to fit aural assay parameters. With little data, the advice is generally added controlled, apple-pie and unique, authoritative it added valuable.
Misconception #4: Big abstracts is alone for big businesses.
Big-data technologies are no best acutely expensive. Companies able-bodied out of the Fortune 500 are application big data. It's not aloof for assertive industries, either; there are big-data technologies geared against about every industry, because best organizations, including abate ones, aftermath astronomic amounts of data. One of the key takeaways from a 2011 McKinsey Global Institute address alleged "Big data: The abutting borderland for innovation, competition, and productivity" was this: "The use of big abstracts will become a key base of antagonism and advance for alone firms." The address begin aboriginal examples of big abstracts in every area it advised — and that was in 2011. Anticipate about how the ability of big abstracts and technology has broadcast back then.
Big abstracts isn't as complicated as best bodies think. Sure, best of us will never accept the algorithms that accomplish it possible, but you use big abstracts in your accustomed activity after alike acumen it. How do you anticipate Pandora chooses your abutting song or Netflix selects your recommended shows and movies? That said, it's important to bethink that not aggregate you apprehend about big abstracts is true. Accomplish abiding you don't abatement fool to one of the big-data myths.
Follow all of the Expert Voices issues and debates — and become allotment of the altercation — on Facebook, Twitter and Google+. The angle bidding are those of the columnist and do not necessarily reflect the angle of the publisher. This adaptation of the commodity was originally appear on Live Science.
from Finr.co http://ift.tt/1JO7Oy9
0 Response to "Misconceptions About Big Data You Can Stop Believing"
Post a Comment