Evidence-Based Analysis

A belief-propagation logic network to objectively evaluate hypotheses against supporting or opposing evidence.

🎯 What Problem It Solves

In investigations, competitive analysis, or decision making, confirmation bias can lead to wrong conclusions. This tool maps evidence to analyze hypotheses objectively.

🧠 Thinking Logic & Core Concepts

A hypothesis's confidence is determined by supporting evidence (+) and opposing evidence (-). Tracing and combining these inputs using fuzzy belief propagation computes the final confidence of the hypothesis.

📋 Newbie Step-by-Step Guide

1. State Hypothesis: Put the core claim at the top.
2. Add Evidence: Map out all findings, data points, or observations.
3. Assign Direction: Use green nodes (+) for support and red nodes (-) for opposition.
4. Define Intermediate Inferences: Connect evidence to sub-claims when needed.
5. Propagate Confidence: Adjust evidence confidence and see how it affects the hypothesis.

💡 Classic Example

[Competitor Product Launch Case]

• Hypothesis: Competitor A will launch a new product next quarter.
• Support (+): Hiring hardware QA engineers; new wireless patents registered; secret distributor briefings.
• Opposition (-): No spike in upstream component orders; hardware lead resigned.
• Result: Despite the resignation, patent and hiring indicators yield a 75% launch probability.