AI driven real-time data analytics solution for spatial-temporal multi-feature data

Challenges
Challenges
US Department of Defense seeks to prototype a readily available solution that includes the ability to: Ingest, clean, and fuse spatial-temporal multi-feature sensor and positioning data (comprised of time, location, velocity, direction, and other continuous and discrete attributes) from a large number of diverse sources and in a variety of formats. Present visualizations, summaries, sensor/data-stream status, and other dashboard views as part of the interface. Accurately classify time-series, spatial entities and predict their future behavior. Allow for data to be missing or incomplete, and both synchronous and asynchronous data transfers. Manage missing or incomplete data and make interpolations or predictions to compensate.
Required features
Be capable of receiving, integrating, and storing spatial-temporal multi-feature data from a large number of sources and a variety of formats. Be rapidly expandable to easily integrate additional data sources in the future. Be Internet-of-Things (IOT) ready in order to interact with emerging sensors and capabilities. Ability to develop/assess algorithms for data discovery, enrichment, visualization, publication and decision-support. Include query-able databases to facilitate access and analyses of the data by third-party, user-centric applications via standard APIs. Employ automated data protection and access control mechanisms to ensure responsible access to data, with capabilities like fine-grained attribute-based access control.
Required features
Xen.AI solution
Xen.AI solution
Xen.AI proposed a live streaming data analytics architecture using Apache Spark Streaming, Machine Learning, Deep Learning and other cloud hosted services from Amazon Web Services (AWS). Will use AWS Cloud services, multi region and multi availability zones, Virtual Private Cloud (VPC) architecture for maximum scalability, security, redundancy and availability. Will implement Access Control Lists (ACLs) and end to end data encryption for maximum data protection and security. Will utilize the latest open source technologies like Google TensorFlow, Scikit-learn, Keras, Numpy, Pandas, Apache Spark, SparkML etc.


US Department of Defense (DoD) seeks a flexible, real-time data analytics solution. The solution must be able to ingest hundreds of GBs of spatial-temporal multi-feature data per day, synthesize and fuse the data, accurately classify time-series and spatial entities, and predict their future behavior. Data originates from diverse sources and arrives in a variety of formats. The ability to synthesize and fuse this data is time sensitive.

We can help build innovative solutions and applications using Artificial Intelligence technologies.