Most people today are aware that businesses and infrastructure are increasingly at risk of a broad spectrum of security events, including terrorism, Advanced Persistent Threats (APT), fraud, insider attacks, and other criminal activities. While many are focused on creating technologies to counter such threats, others are developing methods that can proactively track suspicious behavior patterns and activities in order to reveal hidden threats and dangers in enough time to take immediate action to prevent them from occurring in the first place.

One approach, known as Information Fusion, is seeing significant inroads as a means to stop threats before they can be executed. For example, governments and large organizations currently have custom-built, complex solutions to capture, store, and search large amounts of real-time, streaming data from different sensors and sources such as satellites, monitoring systems, communication networks, mobile devices, and digital media. Built on extremely scalable and distributed technologies, these solutions enable analysts to build on top of the data and add human inferences to help create a realistic view of the situation being monitored. This incorporation of human reasoning as part of the data being collected is the differentiation between basic Data Integration and complex Information Fusion.

The value of Information Fusion systems is to provide in-time access to all relevant components of Big Data for analytic applications. This enables the discovery of hidden connections and relationships within disparate data, allowing analysts to “connect-the-dots” and develop a common view of multiple data streams to support advanced analytic applications. Objectivity’s products, Objectivity/DB and InfiniteGraph, are based on object and graph oriented technologies, making them well-suited to support this type of data organization as both entities and complex relationships between these entities in order to natively address the connections within data and easily handle the massively distributed and scalable requirements.

As these solutions take in streaming, real-time data from smart, distributed sensors, they store the different types of raw data (structured, semi-structured, and unstructured) into a data repository by “refining or fusing” this data through a series of processes such as filtering, classification, transformation, dimensionality reduction, and extraction to create a single view of all of the data. Analysts are then able to search and review the information as it is presented and add human intelligence (Humint) for semantic context that is searchable in the form of relationships, connections, and inferences based on other relevant external factors. These humint associations become part of the data, adding value and inferences to future analysis. External analytical tools may be used to access data repositories of real-time and value-added humint data, enabling users to perform many types of analysis for improved decision support in mission critical situations.

Information Fusion takes specialized data integration for real-time sensor data one step further. By adding and saving inferences and detail that can only be derived with human analysis and support into existing and new data, organizations are able to maximize their analytics efforts and access a more complete “Big Picture View” of a situation, increase the value of the data that they have, and respond more proactively to protect people, assets, and information against all types of security attacks.