AetherEdge KNN Breakout Fortress

The AetherEdge KNN Breakout Fortress indicator is a price congestion tool that utilizes K-Means clustering to identify support and resistance "walls" while employing a K-Nearest Neighbors (KNN) classifier to differentiate between genuine breakouts and fake-outs. By treating historical breakouts as training data, the system adapts its probability scoring to specific market conditions and zone characteristics.

Usage

The Usage section describes how the script can be used to identify high-probability trading opportunities. The indicator automatically draws "fortress" zones based on price density; users should monitor the color-coding of these zones to gauge their strength.

• Strong Wall Reversal: Green-colored zones indicate high historical reliability. Traders often look for price rejections or counter-trade opportunities when price tags these "strong" walls.
• High-Conviction Breakouts: When price breaks above or below a green zone with volume confirmation, it signals a statistically significant trend start.
• Fake-out Avoidance: Orange zones represent "weak" walls. Traders use these to avoid entering on breakouts that have a high probability of failing, instead preparing for potential mean-reversion.
• Trend Following: Once a zone is broken and validated, the indicator extends the zone to act as a future reference point for support or resistance flips.

Details

The indicator operates through a multi-stage AI-driven architecture to refine price action analysis.

• K-Means Zone Detection: The tool collects price points (High, Low, Close) within the lookback window and applies volume weighting. It then runs Lloyd’s algorithm to find optimal cluster centroids, which are converted into horizontal zones if they meet the minimum density threshold.
• 7D Feature Extraction: For every detected zone, the script extracts seven distinct features: normalized density, distance from current price (in ATR units), volatility regime, RSI at formation, Bollinger Band width, zone age, and the number of previous touches.
• KNN Fortress Classifier: These features are stored in memory. When a new zone is formed or tested, the KNN algorithm searches for the most similar past samples to calculate a probability score.
• Pending Learning: This is a dynamic update system. When a break occurs, the indicator waits for a "Confirmation Window" (defined in settings). If price continues in the break direction, it is labeled "Real"; if it returns into the zone, it is labeled "Fake." This data is then added to the KNN memory to improve future predictions.

Settings

K-Means Zone Detection

• Lookback Window: The number of historical bars used to calculate price clusters.
• K (zones): The maximum number of price clusters (walls) to attempt to identify.
• K-means Iterations: The number of passes the algorithm makes to refine cluster accuracy.
• Min Cluster Density: The minimum number of price touches required for a cluster to be displayed as a zone.
• Zone Width (× ATR): Controls the vertical thickness of the zones based on market volatility.

KNN Fortress Classifier

• K Neighbors: The number of similar past samples the KNN algorithm examines to determine the current zone's probability.
• Memory Capacity: The total number of past breakout events stored for learning.
• Real Breakout Score: The probability threshold at which a wall is colored as "Strong" (Green).
• Fake Breakout Score: The probability threshold at which a wall is colored as "Weak" (Orange).
• Confirmation Window: The number of bars the script waits after a break to determine if the outcome was successful or a fake-out.

Break Detection

• Break Buffer (× ATR): The distance price must travel beyond a zone to trigger a breakout signal.
• Vol Confirm Multiplier: The required volume surge (relative to the moving average) to validate a breakout.

FAQ

How do I interpret the zone colors? Green zones (Strong) represent levels where the KNN classifier detects a high probability of a successful break or strong rejection. Orange zones (Weak) suggest a high likelihood of fake-outs based on historical similarities. Gray zones are neutral.

Why do the zones and statistics reset? The KNN memory and stats dashboard are calculated locally on your chart. Refreshing the page or changing certain inputs will reset the learning memory, which then requires new data points to rebuild its predictive accuracy.

How can I access the AetherEdge KNN Breakout Fortress? You can get access on the LuxAlgo Library for charting platforms like TradingView, MetaTrader (MT4/MT5), and NinjaTrader for free.