Kingfisher’s VARYSS product is the result of over a decade of experience using Artificial Intelligence (AI), Machine Learning (ML), complex statistical analysis, and modeling of large datasets to find unique and surprising answers to various problems for the Department of Defense (DoD), Department of Homeland Security (DHS), and the Intelligence Community (IC).
Our analysis shows the U.S. retains a significant lead in knowledge relevant to constructing hypersonic weapons. Our Hypersonic Weapons White Paper discusses our analysis and findings.
China (and North Korea)
One of the topics of continuing interest in China is North Korea. Understanding how topics related to North Korea trend over time in China helps to understand Chinese and North Korean bilateral relations.
Iran: Recent Behavior Follows Internal Signaling
For the past 12 years, VARYSS tracked and monitored political, economic, and military activities in Iran, as well as the entire Middle East, with a focus on internal Iranian and regional political interactions. Using a blended measure of cohesion, belligerence, and framing, VARYSS detected an increase in Iranian behavior signifying a desire to engage in riskier geopolitical activity.
North Korea has been a unique challenge for the United States since the fall of the Soviet Union. For more than two decades, none of the countries that have helped shape North Korea has found enough leverage, either on its own or as part of a coalition, to push North Korea to denuclearize, to join the international community, or to liberalize its economy.
Artificial Intelligence and Data Distillation
At Kingfisher, Machine Learning (ML) is viewed as a subset of Artificial Intelligence (AI) where — with our principal partner PermissionBit — there has been a focus on deep learning for time series and cybersecurity. Our approach to Natural Language Processing (NLP), described in a separate NLP White Paper on our website, is another focused component of our overall AI approach.
Natural Language Processing
Kingfisher has deep expertise interfacing Natural Language Processing (NLP) pipelines with core algorithms in Artificial Intelligence, including deep neural networks. Our focus has been extracting actionable quantitative intelligence from open source media. Larger datasets and enhanced tools for topic discovery allow us to discover complex and emergent topics, and the combination of high frequency time series analysis and automated entity and concept discovery provides increased understanding of causes and effects.