BBVA/Tarkin

A tool for anomaly detection over streaming data based on sentiment analysis

36
/ 100
Emerging

This tool helps security analysts quickly identify unusual and potentially malicious activities buried within vast streams of security logs. It takes raw log data as input and highlights messages that exhibit both infrequent character patterns and negative sentiment, presenting them as prioritized alerts. Security operations centers and incident response teams can use this to cut through the noise and focus on critical issues.

No commits in the last 6 months.

Use this if you need to rapidly detect security anomalies in real-time log data by mimicking a human analyst's intuition for spotting trouble.

Not ideal if your primary need is general log aggregation, simple keyword filtering, or if you require a system that explains its anomaly detection with highly detailed, transparent rules.

security-operations incident-response threat-detection log-analysis cybersecurity
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 13 / 25

How are scores calculated?

Stars

30

Forks

5

Language

Jupyter Notebook

License

Apache-2.0

Last pushed

Jul 02, 2018

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/BBVA/Tarkin"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.