SAN FRANCISCO, Sept. 17, 2019 /PRNewswire/ -- Datameer, the world's leading data preparation solution, today announced the general availability of Datameer X, a new release of the company's industry-leading data preparation and exploration software built for data scientists and machine learning engineers. Now, organizations can speed up machine learning analytics cycles and create robust data flows that feed more data into machine learning models to increase their accuracy.
In today's data-driven world, companies are increasingly investing in artificial intelligence and machine learning. IDC predicts that spending on AI will grow to $77.6 billion by 2022, more than tripling from the $24 billion the market realized in 2018.
Yet machine learning analytic cycles remain hindered as data scientists still spend up to 90 percent of their time preparing data to feed their AI and machine learning processes. Making matters worse, many of them aren't using the best data to make their models more effective and accurate. Datameer X solves for these two critical aspects of data preparation for machine learning with automated functions that shape data for machine learning and large-scale data exploration that finds the best data for the problem at hand.
Datameer has long been a leader in these two areas with its unique Smart AnalyticsTM and Visual Explorer features. Datameer X extends this leadership with:
- New point-and-click machine learning encoding functions—including OneHot, Ordinal, and Date and Number Binning—that automate the process of shaping data
- Pivot table capabilities (similar to Excel) that allow users to organize data into tables at scale with support for billions of records, hundreds of attributes, and thousands of categorical values
- A new production execution mode which allows operationalized jobs to run up to 30 percent faster
- Enhanced integration with external data sources and destinations such as Google BigQuery and Tableau Hyper
"The new encoding and pivot table features in Datameer X will truly accelerate data preparation for machine learning initiatives and allow data scientists to focus on what they do best: build robust AI models," says Christian Rodatus, Datameer CEO. "The product's point-and-click functionality lets data scientists find the best data and feed their models 10 to 20 times faster, eliminating errors along the way."
Press Contact: Benoite Yver, [email protected]