SAN FRANCISCO—At Microsoft's Build conference here today, corporate vice president of Microsoft's machine learning unit Josephe Sirosh discussed some of the applications already leveraging data analytics and machine learning services in Microsoft's Azure cloud. Among the early adopters: Japanese cows.
In 2013, Fujitsu introducedGyuHo SaaS, a cloud-based system for dairy farmers that helps track the health of their herds through Wi-Fi connected pedometers—essentially giant Fitbits for cows. The time and movement data can help farmers not only track the general health of cattle, but can also help track when cows are going into estrus (a condition more commonly known as "heat").
Fujitsu built the data analytics for GyuHo (which is Japanese for "cow step") in the Azure cloud. Using Azure machine learning logic, the software-as-a-service application can detect spikes in movement activity at night that are an indicator that a cow is going into estrus and is ready for artificial insemination. Sirosh said that by alerting farmers when data suggested estrus was beginning (which the system can do with 95 percent accuracy), they could raise their successful insemination rate from about 30 percent, based on daily hands-on cow inspection, to 65 percent.
Read 3 remaining paragraphs | Comments
No comments:
Post a Comment