Understanding Trust Scores

Learn how TRAST's dynamic trust scores work, incorporating reputation-weighted feedback, time decay, and AI-driven anomaly detection for reliable Web3 entity evaluation.

TRAST utilizes a dynamic and nuanced trust scoring system designed to provide a reliable assessment of Web3 entities (projects, tokens, profiles, etc.). Scores are represented visually with a 5-square system and a percentage (0-100%).

📊 Core Components & Display

Each entity display shows:

[Score Squares] [Percentage]% • 👥[Reviews] • 👀[Views]

Example: 🟩🟩🟩🟨🟨 82% • 👥45 • 👀120

• Score Squares: Quick visual indicator (🟩 High, 🟨 Medium, 🟥 Low Trust).

• Percentage: Precise numerical score based on aggregated feedback.

• Reviews (👥): Number of unique users who have rated/reviewed.

• Views (👀): Indicates community interest and visibility.

Key Features of the Scoring System (v0.1-beta onwards)

TRAST's scoring goes beyond simple averages:

1. Reputation-Weighted Feedback:

   • Reviews and ratings from users with higher trust levels and verified status (✅) have a greater impact on the score.

   • Anonymous (👻) or new users (🌱) still contribute, but with less weight initially.

   • This mitigates manipulation and prioritizes input from established community members.

2. Time-Based Decay:

   • The influence of older reviews gradually decreases over time.

   • This ensures scores reflect the current community sentiment and entity status, not just historical data.

   • Recent activity and feedback have a stronger impact.

3. AI-Driven Anomaly Detection (Scam Radar integration):

   • The system uses AI (Scam Radar) to monitor for suspicious rating patterns (e.g., coordinated voting, sudden spikes).

   • Anomalous activities can trigger alerts or adjustments to maintain score integrity.

   • Helps protect against manipulation attempts.

4. Multi-Factor Analysis:

   • Scores consider not just ratings but also the number of reviews, review quality (via AI analysis - future), entity verification status, and other signals.

🏆 Impact on Trending & Visibility

Entities are ranked in trending lists and search results based on:

• Trust score percentage

• Number and quality of reviews

• View count and recent activity

• Verification status

Higher, well-supported trust scores lead to greater visibility.

💡 Using Scores Effectively

• Look Beyond the Number: Check the number of reviews (👥). A high score with few reviews is less reliable.

• Consider Recency: Prioritize recent feedback, especially due to time decay.

• Read Comments: Use the Commenting System to understand the context behind ratings.
• Note Verification: Verified entities (✅) generally indicate a higher level of transparency.

• Cross-Reference: Use TRAST scores as a vital data point alongside your own research (DYOR).

🛡️ Safety Tips

• Red Flags: Sudden score drops, very few reviews despite high views, overwhelmingly positive/negative reviews without substance, community warnings in comments.

• Best Practices: Use scores as guidance, read detailed reviews and comments, check multiple sources, report suspicious activity.

TRAST trust scores provide a dynamic, community-powered layer of intelligence for navigating the Web3 space.