Splunk

Splunk

Behavior detection product "SplunkUBA" by machine learning

Splunk User Behavior Analytics (hereafter, Splunk UBA) is an out-of-the-Box (ready-to-use) solution that detects unknown threats and anomalous behaviors through machine learning.
Splunk UBA generates a baseline from the input data using machine learning, detects anomalies based on the baseline, performs further machine learning on the generated anomalies, and detects threats. increase.
It also provides visibility into unknown threats and multiple entities (users, devices, apps) related to threats to understand the overall security story. If it is determined to be a threat, Splunk products will conduct more detailed investigation and countermeasures.

Are you having trouble with these issues? : Internal fraud

I want to detect information taken out by malicious employees
Individuals who transfer files to multiple overseas cloud services, etc.
Identify individuals that are difficult to detect on a threshold basis!

Are you having trouble with these issues? : Countermeasures against targeted attacks

I want to detect advanced security attacks
You can't notice it just by responding to rule-based security alerts
An attempt to log in to an important internal server using a hijacked account has been detected!

screen image

TOP screen

Threat details screen

feature

  1. Multiple analysis with anomalies and threats
    Exposing anomalies in behavior using machine learning, and extracting threats from those anomalies
    In the case of anomaly detection alone, a large amount of anomalies are detected, but by multiplexing machine learning, it is possible to extract highly accurate threat information.
    By detecting threats that operators really need to see and receiving advice on countermeasures, operators can avoid alert fatigue (Alert Fatigue/Alarm Fatigue).

    Alerts for anomalies and scoring only

    Too many alerts obscure real threats, what to look for
    Also, if you do not know what is detected, you do not know how to deal with it
    Splunk UBA scores anomalies and threats separately, so the real threat is clear.
    In addition, there is an explanation of what the threat is and how to deal with it.
    This reduces the burden on operators after detection.
  2. Multi-entity machine learning model
    Detect anomalies and threats that combine not only user behavior but also device and application behavior

  3. unsupervised machine learning
    Detect anomalies/threats based on historical data without preparing correct information
  4. User-friendly interface
    Detailed descriptions of threats and anomalies are available, allowing for quick decisions on next actions
  5. Seamless integration with Splunk products
    Functional affinity such as data collection using Splunk and feedback of detection events to ES
  6. Real-time/big data platform
    Scalability through scale-out

Configuration image

Inquiry/Document request

In charge of Macnica Splunk Co., Ltd.

Mon-Fri 8:45-17:30