Data science startup Explorium Ltd. wants to play matchmaker for the enterprise after landing $19 million in funding to get its operation off the ground.

The funding, announced today, consists of a $3.6 million seed round led by Emerge with participation from F2 Capital, plus a $15.5 million Series A round led by Zeev Ventures.

Explorium’s idea is to provide enterprises with richer data sets that can be used to train machine learning models, which can then be used to make more accurate predictions regarding aspects of their businesses.

Most companies already have stacks of data stored within their internal information technology systems, but machine learning models become a lot more reliable if they can be fed with additional third-party data. That’s where Explorium comes into play, helping companies connect their data to thousands of other publicly available information sources.

Explorium bills its platform as a kind of “dating site” for data. It works by analyzing a company’s data to work out which additional information can improve their machine learning models. This new data is then delivered to customers in a ready-to-use format.

Explorium’s cofounder and chief executive officer Maor Shlomo told SiliconANGLE that the external data his company provides adds critical context to organization’s existing information, thereby enabling more effective machine learning.

“For example, an insurance provider is attempting to predict how many calls to expect in any given time in order to effectively staff their call center,” Shlomo said. “However, their historical dataset consists mainly of just the date, time and number of calls received in the past. This is too narrow to provide a clear picture of the underlying dynamics, resulting in an inaccurate model that you simply can’t trust with running your business.”

Shlomo said Explorium would be able to enhance the insurance providers’ data sets by exploring things such as the time the calls were made. Then it would investigate if, for example, it was raining at the time of the calls, whether or not it was dark outside, and if there were any news reports of damages.

Explorium’s platform explores thousands of auto-generated ideas to enhance customer’s data in this way, enabling it to provide more context. With that added context, the insurance provider would be able to make more accurate predictions on when it’s likely to receive a deluge of calls.

“We are doing for machine learning data what search engines did for the web,” Shlomo said. “Just as a search engine scours the web and pulls in the most relevant answers for your need, Explorium scours data sources inside and outside your organization to generate the features that drive accurate models.”

Although it’s only just emerging from stealth mode now, Explorium already has more than 60 employees and counts several companies from the Fortune 100 as its customers.

Image: geralt/Pixabay