## Flowchart: Factual Accuracy Evaluation Process
### Overview
The image displays a simple, two-step decision flowchart designed to evaluate the factual accuracy of a piece of information or a statement. The process uses two sequential yes/no questions to categorize the input into one of three outcomes: Accurate, Minor Inaccurate, or Major Inaccurate.
### Components/Axes
The flowchart consists of the following elements, connected by directional arrows indicating the flow of the decision process:
1. **Decision Diamond 1 (Top-Left):**
* **Text:** "Is it related to the context"
* **Shape:** Pink diamond.
* **Outputs:** Two arrows labeled "No" (pointing right) and "Yes" (pointing down).
2. **Outcome Rectangle 1 (Top-Right):**
* **Text:** "Major Inaccurate (Non-factual 1)"
* **Shape:** White rectangle with a black border.
* **Input:** The "No" arrow from Decision Diamond 1.
3. **Decision Diamond 2 (Center-Left):**
* **Text:** "Is it Factual? e.g. using Wikipedia / Google Search"
* **Shape:** Pink diamond.
* **Input:** The "Yes" arrow from Decision Diamond 1.
* **Outputs:** Two arrows labeled "No" (pointing right) and "Yes" (pointing down).
4. **Outcome Rectangle 2 (Center-Right):**
* **Text:** "Minor Inaccurate (Non-factual 0.5)"
* **Shape:** White rectangle with a black border.
* **Input:** The "No" arrow from Decision Diamond 2.
5. **Outcome Rectangle 3 (Bottom-Left):**
* **Text:** "Accurate (Factual 0)"
* **Shape:** White rectangle with a black border.
* **Input:** The "Yes" arrow from Decision Diamond 2.
### Detailed Analysis
The flowchart defines a clear, hierarchical logic:
1. **Primary Filter (Context Relevance):** The first gate checks for contextual relevance. Information deemed unrelated to the context is immediately classified as "Major Inaccurate" and assigned a severity score of `1`.
2. **Secondary Filter (Factual Verification):** If the information is contextually relevant, it proceeds to a factual check. The flowchart suggests using external verification sources like Wikipedia or Google Search.
* If the information cannot be verified as factual, it is classified as "Minor Inaccurate" with a severity score of `0.5`.
* If it is verified as factual, it is classified as "Accurate" with a severity score of `0`.
### Key Observations
* **Severity Scoring:** The outcomes include a numerical score in parentheses (`1`, `0.5`, `0`), suggesting a quantitative system for measuring inaccuracy. "Major Inaccurate" is penalized twice as heavily as "Minor Inaccurate."
* **Visual Coding:** Decision points are consistently represented by pink diamonds, while terminal outcomes are in white rectangles.
* **Linear Flow:** The process is strictly linear and top-down, with no loops or feedback mechanisms. A piece of information can only be evaluated once.
### Interpretation
This flowchart presents a simplified model for triaging information accuracy, likely intended for content moderation, fact-checking, or research validation. It prioritizes **contextual relevance as the primary determinant of severity**. An irrelevant claim is considered more damaging (`1`) than a relevant but unverifiable one (`0.5`).
The model implies that the most critical failure is not necessarily being wrong, but being wrong *and* off-topic. The "Accurate" state requires passing both filters: relevance and verifiable factuality. The inclusion of specific examples ("Wikipedia / Google Search") grounds the abstract concept of "factual" in common, accessible verification practices. This tool would be useful for quickly categorizing inputs in a systematic way, though its simplicity may overlook nuances like the credibility of sources or the intent behind the information.
