MarkLogic partners with DoD and JAIC to structure data for national security

It’s all about the metadata.

When preparing unstructured data for AI, you’re really pulling in all the data about that data, said Jeffrey Casale. And this, noted the CEO of data integration and management company MarkLogic, is ultimately what proves to be the most useful. It provides context, insight, meaning and key messages.

[aditude-amp id="flyingcarpet" targeting='{"env":"staging","page_type":"article","post_id":2791968,"post_type":"story","post_chan":"ai","tags":"category-business-industrial,category-computers-electronics-enterprise-technology","ai":true,"category":"ai","all_categories":"ai,big-data,data-infrastructure,","session":"C"}']

“We can understand not just the information, but information about the information,” Casale said. “And that allows us to build powerful models.”

As organizations ingest more and more data by the day – and increasingly, in some cases, by the moment – they are looking to AI and ML for help. MarkLogic has developed its enterprise NoSQL platform as a means to support organizations in managing, exploiting and securing their data – and soon this tool will be leveraged for U.S. national security purposes.

Data for defense

The 20-year-old, San Carlos, California-based company is preparing to work with the Department of Defense (DoD) as part of a $241 million basic ordering agreement. This supports the Joint Artificial Intelligence Center’s (JAIC) Data Readiness AI Development program, which is designed to scale AI across the Pentagon.

MarkLogic is one of several dozen organizations chosen to partner with JAIC, which was established in 2018 with a charge to “seize upon the transformative potential of artificial intelligence technology for the benefit of America’s national security.”

Casale couldn’t provide specific details on the work to be done or its scope as it is still very early in the process, he explained — and the work will ultimately be classified. The Basic Ordering Agreement stipulates that MarkLogic will work with the JAIC to identify appropriate AI use cases, develop capabilities and scale data impacts.

“How do we allow organizations within the DoD to take advantage of the data they currently have and allow that data to be more effective in solving complex challenges that the government faces on any particular mission?” Casale said.

According to its website, the JAIC is applying AI in areas of joint war fighting (systems, sensors and targeting solutions), war fighter health (disease curing, decreasing healthcare costs and increasing operational readiness), joint information warfare and logistics. It is also leveraging AI to improve fleet readiness through diagnostics, training, process improvements, forecasting and supply chain optimization — and to transform mission areas that are aligned against unpredictable threats such as humanitarian resistance and disaster relief, countering weapons of mass destruction and force protection.

Whatever the specific use cases, MarkLogic is looking forward to the opportunity to help “solve the nation’s most critical national security complex data challenges,” Casale said.

[aditude-amp id="medium1" targeting='{"env":"staging","page_type":"article","post_id":2791968,"post_type":"story","post_chan":"ai","tags":"category-business-industrial,category-computers-electronics-enterprise-technology","ai":true,"category":"ai","all_categories":"ai,big-data,data-infrastructure,","session":"C"}']

“We are very proud to have won the recognition of JAIC to support their data readiness efforts and we look forward to continuing our commitment to supporting the ethical principles for AI. DOD’s award makes MarkLogic a strategic partner with JAIC.”

Expanding use in the federal sector

The contract award expands on MarkLogic’s vast experience in the federal sector. Its platform has been used in significant projects for the U.S. Marine Corps and the Center for Tobacco Products’ Integrated Research Data System (CIRDS) and in the implementation of Healthcare.gov.

The U.S. Air Force Research Lab, meanwhile, leveraged MarkLogic technology to create its HyperThought data management platform. This provides a scalable, agile and flexible way to make data at the exabyte level discoverable and securely shareable with thousands of internal and external scientists and engineers. According to the agency, it has increased performance a hundredfold in addition to saving tens of thousands of development hours.

Data agility for the enterprise

In enterprise, meanwhile, Boeing has used the MarkLogic platform for integrating data to enable digital twins; Aetna has created a human resources hub and pharmaceutical R&D company AbbVie has powered its PubLab. This tool uses ML, natural language processing, visualization and analytics to allow data scientists to search and discover more than 40 million pieces of unstructured and semi-structured scientific literature.

[aditude-amp id="medium2" targeting='{"env":"staging","page_type":"article","post_id":2791968,"post_type":"story","post_chan":"ai","tags":"category-business-industrial,category-computers-electronics-enterprise-technology","ai":true,"category":"ai","all_categories":"ai,big-data,data-infrastructure,","session":"C"}']

“The challenge is that organizations have vast amounts of data,” Casale said. And they all want to pull meaning from it, but “the more complex the data, the harder it is to get at the data.”

The MarkLogic platform has been used for deep search and query, to build enterprise applications and to bring insights to analytics and ML, he explained. The platform’s advantage is that it ingests data as-is from any source and stores that data – along with everything known about it – as a single resource in a unified, secure platform.

This connects and feeds AI systems and enables data to be tracked and traced throughout the algorithm, Casale said. Ultimately, this delivers high-quality, trusted, accessible insights that drive rapid action and innovation and it is especially helpful when reviewing why decisions were made or why future actions should be taken.

“It’s the unique ability that we have with complex data,” Casale said. “We can develop powerful models that really establish this data agility and security. It really is only as good as your ability to get meaning out of all the data that you already have.”

[aditude-amp id="medium3" targeting='{"env":"staging","page_type":"article","post_id":2791968,"post_type":"story","post_chan":"ai","tags":"category-business-industrial,category-computers-electronics-enterprise-technology","ai":true,"category":"ai","all_categories":"ai,big-data,data-infrastructure,","session":"C"}']