LUX interprets tens of billions of data points daily.

Your infinite possibilities are funneled to the most critical, usable intelligence.

  • Data Ingest

    LUX can import structured and unstructured data from hundreds of sources, both public and private.

  • Enrichment

    Information can be scored, validated, translated, or summarized using AI/ML techniques.

  • Analysis

    Hundreds of algorithms can be applied to identify anomalies or predict future behavior.

  • Correlation

    Powerful searching, filtering, and matching capabilities to identify critical actionable information.

  • Alerting

    Inform key individuals or other platforms with most valuable insights.

Abstract Image a Pipeline

LUX empowers analysts

LUX conducts real-time processing of fused data sources using a combination of machine learning, streaming analytics and complex event processing to match the results against questions posed by analyst teams.

LUX Analysts Diagram LUX can ingest nearly any kind of data, then conduct real-time processing of enriched data sources using a combination of AI, Machine Learning and Streaming Analytics to match results against user questions. When a match occurs, a report is generated and sent to the analysts where they can use LUX visualizations to further interrogate and understand the importance and relevance of the alerts.

Platform Features

  • All Data

    LUX works with virtually any data type, including real-time and persistent, numeric or text! New data sources can be added in a matter of hours.

  • Open Architecture

    LUX can communicate with most analytical and AI platforms programmatically (APIs) to leverage the appropriate tools to accelerate the discovery process.

  • No-code Interface

    Any non-technique business analyst can easily pose LUX questions until the answers are identified. More complex analytic engines can be developed as well.

  • Secure and Scalable

    LUX has been certified for use on the U.S. government’s highest classified networks. Our analytic engines can process millions of real-time or persistent records (events) in minutes!

  • Visualization

    LUX and other 3rd party visualizations are utilized to interrogate and understand the importance of enriched data results.

  • Workflow Automation

    Enables automated alerts of summarized content at any stage in the data funnel to key individuals, groups, or your favorite workflow products.

Our no-code software solution.

LUX enables domain experts to ask questions in a highly iterative fashion without the need to involve software developers. LUX also allows for more complex algorithms to be envisioned and implemented to further accelerate your discovery process.

Analytics visualizations displayed on three smartphones

Deployment Options

LUX Technology Stack

UX
  • Angular
  • Chart.js
  • OpenLayers
  • Cytoscape
Backend
  • Apache Spark
  • Spring Framework
  • jQuery
  • Apache CXF
  • Deeplearning4j
  • OpenCV
Infrastructure
  • Java
  • MongoDB
  • TensorFlow
  • PyTorch
  • Python
  • Apache Tomcat
Operating System
  • CentOS
  • Android
  • Red Hat
  • Ubuntu
  • OpenWrt
  • Windows
Architecture
  • x86 Architecture
  • ARM Architecture

Ingest Plugins

  • Twitter
  • Facebook
  • Reddit
  • Wikipedia
  • Mastodon
  • Gab.ai
  • WT.Social
  • Pastebin
  • YouTube
  • Spinn3r
  • Webscraper
  • News - RSS
  • News - GDELT
  • News - Bright Planet
  • News - Reuters
  • News - AP
  • Darkweb - Darkowl
  • Bitcoin
  • OpenSky FAA
  • Pinkfroot FAA & AIS
  • NOAA Earthquake
  • NOAA Buoy
  • WMATA Buses
  • Commercial AIS
  • NMAP
  • 802.11
  • Mobile Ad-Tech
  • PreData

Analytics Plugins

  • Named Entity Recognition
  • Translation
  • Emotion Detection
  • Sentiment Detection
  • Topic Detection
  • Term Trend Analysis
  • Emerging Topics
  • Wordclouds
  • Image OCR
  • Motivational Analytics

AI/ML Analytical Techniques

  • Abnormal Track Analytic

    • KMeans track classifiers
    • Naive Bayes classification
  • Geo Grid Track Forecast Analytic

    • In-house “Geospatial neural network”
  • Topic Discovery

    • In-house ML pipeline called “RT2”
    • LDA for keyword extraction
    • fastText (Meta) for word vectorization
    • ICG KMeans (below)
    • ChatGPT Integration
  • Track Gap Proximity

    • A* pathfinding, uses in-house real-time
    • worldwide lookup data model w/ nanosecond
    • performance for obstacle avoidance
  • Indicators Analytic

    • Probabilistic decision making with pluggable
    • aggregation algorithms
  • Normalcy Analytics

    • Leans multivariate patterns
  • Sentiment/Ethnicity/Gender Enrichments

    • NLP
  • Word2Vec Enrichment

    • Neural network to categorize text
  • OCR Enrichment

    • Extracts text from imagery using computer
    • vision
  • Recognition Enrichment

    • Amazon API for object/celebrity labeling for
    • images
  • NLP Enrichment

    • Natural Language Processing
    • Named Entity Extraction
    • ChatGPT Integration
  • LDA Enrichment

    • LDA for keyword extraction
  • Entity Behavior Enrichment

    • Expert system / state machine
  • Translation Enrichment

    • Real time text translation
    • ChatGPT Integration
  • Speech to Text filter

    • Transcribes audio to text using DeepSpeech
    • RNN
  • Behaviors Analytic

    • Entity-based state machine
  • ICG KMeans

    • KMeans++ initialization
    • Automated optimal cluster assignment using
    • automated elbow detection
  • HBOS Anomaly Enrichment

    • Histogram-based outlier detection
  • TFLite Classification Filter

    • Classifies images using a pluggable TensorFlow
    • lite model
  • CNN Classifier Enrichment

    • Uses a CNN to classify text
  • Mobile Net Classification Filter

    • Image classification using Mobile Net neural
    • network
  • Graph Clusters Analytic

    • Graph cluster detection