thatDot Novelty

Real-time Anomaly Detection Using Categorical Data

Existing anomaly detection techniques rely on numerical data and threshold analysis, which breaks down in the face of high data dimensionality and produces high volumes of false-positives. thatDot Novelty Detector uses categorical data to build a comprehensive behavioral fingerprint of your data. This deep contextual understanding eliminates false-positives and provides WHY an anomaly was identified, making it immediately actionable.

Fewer False Positives

1000x Scaling Performance

thatDot Novelty Detector uses the categorical data in logs and events to build a rich behavioral fingerprint. Yellow dots with low scores (lower on the plot), are examples of unique data that are not anomalous. “New” is not always “novel”. Only thatDot Novelty Detector automatically uses context to learn the difference.
Novelty Detector is a new A.I. technique built on thatDot's Quine streaming graph. Novelty Detector allows for real-time use cases not possible with existing anomaly detection solutions. The performance comparison was made on identical 8-core, 16 GB, VMs. Note the different left and right hand scales.

Delivering value for real-time use cases

“thatDot’s Novelty Detector for categorical data is the future of anomaly detection. Its speed powers our new real-time services while also significantly reducing false-positive findings for our customers.”
Gery Szlobodnyik
CEO, Traceriser

Simple to Deploy and Use

thatDot Novelty Detector is an easily deployed, highly scalable application. Spin up an AWS or on-premise instance in minutes.

An A.I. Application built on Quine

  • On-premise or cloud installed .jar file or container (not SaaS)
  • APIs for streaming or batch data ingestion
  • Self-training operation, no data labeling
  • Built in data transformation functions
  • A single 8-core instance scales to 15,000 observations/second
  • Integrated and online self-service documentation

Real-time, Actionable Insights

  • Real-time individual or bulk anomaly detection at scale
  • Build a detailed behavioral fingerprint for deep contextual understanding that eliminates false-positives
  • Every observation response includes “why” an anomaly was identified
  • Detailed scores for novelty, uniqueness, information content, and probability
  • Integrate into your code to drive real-time remediation workflows

Delivering value for real-time use cases

Cyber Security

Monitor SaaS and IaaS usage for unusual configuration changes or resource access patterns.

Network Optimization

Identify network route inefficiencies and eliminate redundant alerts through topology awareness.

Fraud Detection

Analyze usage for excess concurrent usage, and generate events to enforce entitlement compliance.

Log Reduction

Actively filter logs and events for to identify the logs worth analysis, and the bulk that are not.

Need To Transform And Join Data First?

Many of the most valuable observations are the result of transforming, combining and interpreting data from many different sources. thatDot's Quine streamlines the process of real-time data cleaning and transformation, including the multiplexing of log, event and metrics data into any desired data format. And, with all thatDot solutions, you can use non-numeric data in it is native format, no conversions to numeric values needed.