Skip to main content

Update frequency & Quality checks

Score update frequency​

Timing matters.

Initially, for simplicity's sake, we could use tools that integrate data, but the final objective is to stay as close as possible to the real-time data sources.

QQ Omega work with different frequencies depending on the type of data:

  • On-chain data → near real-time updates
  • Market data (price, volume, orderbooks) → intraday / minute-level.
  • Social & narrative signals → frequent pulls (every few minutes / hours).
  • Off-chain / fundamentals (team, docs, legal) → (daily / weekly).

Keeping the pipeline fresh, robust and close to the source is a core priority.
If a data source degrades in quality or latency, it is reviewed or replaced.

Quality checks and anti-spam filters​

To keep scores clean and trustworthy, QQ applies multiple safeguards:

  • Data validation
    • Check for missing values, impossible numbers, duplicated entries.
  • Outlier detection
    • Flag suspicious spikes (e.g. fake volume, wash trading, bot-driven social spam).
  • Source cross-checking
    • Compare information across multiple providers when possible.
  • Anti-gaming heuristics
    • Detect patterns of behaviour aimed only at “farming” a better score (e.g. short-term liquidity injections, wash activity to look active).

Only after passing these checks does data flow into the agents and affect scores.
If something looks wrong, data can be flagged, downweighted or excluded.