Search

When a sufficient amount of training data is available, Machine Learning (ML) models show great promise for solving problems involving complex and dynamic patterns. Social and behavioral domains are rich with such challenging problems, with complex object data extracted from documents, surveys, etc., and represented in forms such as graphs and trees. However, many social and behavioral data sets are inherently sparse and incomplete. The same data field may be unavailable in different records of a data set due to different causes, e.g., because it was not measured, not known, or simply not applicable to that particular record. Furthermore, collection challenges, cost, lack of participation, small affected populations, etc., result in very small sets of data. Resulting unconventional datasets cannot be directly used with potent approaches such as machine learning. A technique to model and synthesize large sets of such complex data objects while maintaining the same statistical and topological characteristics of original data helps overcome these challenges. We propose a novel feature mapping technique to eliminate data inconsistencies and model data objects from unconventional datasets. The feature-mapped data objects are used to synthesize data using two likelihood approaches, i.e., multi-variate Gaussian and regular vine copulas, and one generative adversarial approach using an adversarial autoencoder (AAE). We demonstrate the robustness of the proposed technique with three real-world datasets representing disparate domains and validate the performance of likelihood and deep-generative approaches with these object synthesis strategies.

Communications resiliency and data distribution assurance has been compromised in highly competitive, dynamic, and stressing environments with the introduction of off the shelf, high performing jammers. This paper introduces a novel cognitive communications system that can be fielded on small attributable platforms; the architecture amalgamates a highly capable environmentally perceptive aperture, a software defined radio, and sophisticated networking techniques. The proposed cognitive communications system uses the 5G new radio waveform and applies groundbreaking machine learning methods to facilitate systems orchestration amongst its subsystems to transfer information effectively between nodes, and across large-scale multi-hop networks, essential for rapid strike missions. With the challenges imposed by mature and readily available jammers, a cognitive communications system can be used to maintain and sustain continuous communications to provide near real-time surveillance and situational awareness updates. In addition to describing the comprehensive cognitive communications system architecture, a jamming analysis will be presented. Performance results for an operational use case will be compared to existing trite architectures consisting of siloed apertures and standard radio systems to demonstrate an improvement of approximately 10 dB in communications margin with the auspicious technology offering of the cognitive communications system for low size, weight, and power attributable platforms.

Tracking individuals or groups based on their hidden and/or emergent behaviors is an indispensable task in homeland security, mental health evaluation, and consumer analytics. On-line and off-line communication patterns, behavior profiles and social relationships form complex dynamic evolving knowledge graphs. Investigative search involves capturing and mining such large-scale knowledge graphs for emergent profiles of interest. While graph databases facilitate efficient and scalable operations on complex heterogeneous graphs, dealing with incomplete, missing and/or inconsistent information and need for adaptive querying pose major challenges. We address these by proposing an inexact graph pattern matching method, which is implemented in a graph database with a scoring mechanism that helps identify hidden behavioral patterns. PINGS (Procedures for INvestigative Graph Search), a graph database library of procedures for investigative graph search is presented. Results presented demonstrate the capability of detecting individuals/groups meeting query criteria as well as the iterative query performance in graph databases. We evaluate our approach on three datasets: a synthetically generated radicalization dataset, a publicly available patient’s ICU hospitalization stays dataset, and a crime dataset. These varied datasets demonstrate the wide-range applicability and the enhanced effectiveness of observing suspicious or latent trends in investigative domains. [ABSTRACT FROM AUTHOR]

For many important network types, physical coordinate systems and physical distances are either difficult to discern or inapplicable. Accordingly, coordinate systems and characterizations based on hop-distance measurements, such as Topology Preserving Maps (TPMs) and Virtual-Coordinate (VC) systems are attractive alternatives to geographic coordinates for many network algorithms. We present an approach to recover geometric and topological properties of a network with a small set of distance measurements. The approach is a combination of shortest path (often called geodesic) recovery concepts and low-rank matrix completion, generalized to the case of hop-distances in graphs. Results for sensor networks embedded in 2-D and 3-D spaces as well as for social networks indicate that the method can accurately capture the network connectivity with a small set of measurements. TPM generation can now be based on various context appropriate measurements or VC systems instead of distances to a set of global anchors. The proposed method is a significant generalization that allows the topology to be extracted from a random set of graph shortest paths, making it applicable in contexts such as social networks where VC generation may not be possible. [ABSTRACT FROM AUTHOR]