Evidence Fusion Ledger: Bayesian Log-Odds Waterfall |…

What is the Evidence Fusion Ledger?

The Evidence Fusion Ledger is a visualization tool used in cybersecurity for combining various pieces of evidence (often referred to as Open Source Intelligence or OSINT) to form a cohesive and reliable conclusion. It uses Bayesian statistics, which allows it to update probabilities based on new data.

At its core, the ledger tracks each piece of evidence's contribution through a log-odds waterfall chart, where each signal is weighted according to its reliability and then combined to produce a final trust score.

How Does It Work?

The process begins with individual signals (evidence) being converted into their respective log-odds values. These values are then weighted based on the quality of the data, which can be adjusted using the data-quality field in the controls.

As new claims or evidence is added through the 🔄 button, these contributions accumulate on a waterfall chart, visually representing how each piece of evidence impacts the overall trust score.

live demo · related simulation● LIVE

Why Use Bayesian Statistics?

Bayesian statistics are particularly useful in cybersecurity because they allow for the updating of probabilities based on new data. This is crucial when dealing with dynamic and evolving threats, as it enables continuous refinement of risk assessments.

The use of log-odds values helps to normalize and compare different types of evidence, making the fusion process more robust and reliable.

Real-World Applications

In practice, the Evidence Fusion Ledger can be used by cybersecurity analysts to assess the risk associated with potential threats. By combining multiple sources of information, it helps in making informed decisions about whether a threat is credible.

This tool is also valuable for identifying patterns and trends over time, which can aid in proactive security measures.

Frequently asked questions

What does the trust threshold field do?

The trust threshold field sets the minimum confidence level required to consider a conclusion valid. If the final trust score falls below this threshold, it indicates that more evidence is needed before making a decision.

How does adjusting the data quality affect the ledger?

Adjusting the data quality field changes how much weight each piece of evidence carries in the fusion process. Higher values indicate better-quality data and thus have a greater impact on the final trust score.

Can this tool be used for other types of analysis besides cybersecurity?

Yes, while designed for cybersecurity, the Evidence Fusion Ledger can be adapted for use in other fields requiring evidence fusion, such as finance or environmental monitoring.

What does the sigmoid inspector curve represent?

The sigmoid inspector curve represents the transformation of log-odds values into a probability scale. It shows how the trust score is translated from its raw form to a more interpretable format, often between 0 and 1.

Try it live

Everything above runs in your browser — open Evidence Fusion Ledger: Bayesian Log-Odds Waterfall and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.

▶ Open Evidence Fusion Ledger: Bayesian Log-Odds Waterfall simulation