Big data identifies the best bull in America


The big data phenomenon is at work everywhere, including in the age-old dairy cow business. Predictive analytics has caused a "massive transformation" in the dairy herds in the United States, reports The Atlantic's Alexis Madrigal in a lengthy article that explains everything you ever wanted to know about dairy cow breeding, and then some.

Dairy cow breeders are far from newcomers to data analytics. Highly suited to quantitative analysis, dairy breeding involves pedigree records and genetic data that goes back decades. What's more, the traits breeders are interested in--such as milk production, milk content, cow longevity--are few in number and easily measurable. As it turns out, cows only work for three to four years, so smart breeders make good use of genetics.

In recent years there has been a change in what is emphasized in determining a valuable breeder because cows produce a great deal more milk than they used to, while their numbers have decreased considerably. In the past few decades, the size of herds has grown while the number of dairies has dropped dramatically. The business is radically more efficient than it was just a couple generations ago.

"Thousands of years of qualitative breeding on family-run farms begat cows producing a few thousand pounds of milk in their lifetimes; a mere 70 years of quantitative breeding optimized to suit corporate imperatives quadrupled what all previous civilization had accomplished," Madrigal writes. "And the crazy thing is, we're at the cusp of a new era in which genomic data starts to compress the cycle of trait improvement, accelerating our path towards the perfect milk-production machine, also known as the Holstein dairy cow."

For more:
- see Alexis Madrigal's article in The Atlantic

Related Articles:
How big data resembles a jigsaw puzzle
Allstate uses data science contest to improve risk prediction
Big data or just big hype?