{
  "id": 10501,
  "rank": 614,
  "name": "News Sentiment Filter",
  "aka": [
    "Integrating News Sentiment to Avoid Strategy Disruption in MQL5"
  ],
  "family": "B",
  "subtype": "B2",
  "one_liner": "Uses real-time news sentiment to filter trades and exits.",
  "market_regime": "any",
  "timeframes": [
    "M15"
  ],
  "symbols": [
    "EURUSD"
  ],
  "sessions": [
    "any"
  ],
  "indicators": [
    {
      "name": "EMA",
      "params": {
        "period": 20
      },
      "applied_to": "close"
    },
    {
      "name": "EMA",
      "params": {
        "period": 40
      },
      "applied_to": "close"
    },
    {
      "name": "RSI",
      "params": {
        "period": 7
      },
      "applied_to": "close"
    }
  ],
  "entry_long": [
    "Fast EMA crosses above Slow EMA",
    "News sentiment is bullish with confidence >= 0.7"
  ],
  "entry_short": [
    "Fast EMA crosses below Slow EMA",
    "News sentiment is bearish with confidence >= 0.7"
  ],
  "exit_rules": [
    "Position is long and news sentiment is bearish with confidence >= 0.55",
    "Position is short and news sentiment is bullish with confidence >= 0.55"
  ],
  "filters": [
    "news filter"
  ],
  "stop_loss": "SL is placed at price * (1 - slPricePerc / 100)",
  "take_profit": "TP is placed at price * (1 + tpPricePerc / 100)",
  "position_size": "Lots = inpLots",
  "evidence_type": "source_code",
  "completeness": 80.0,
  "signal_warnings": null,
  "claim_warnings": null,
  "opacity_warnings": [
    "PARTLY OPAQUE ENTRY: 2.0 of 4 entry conditions reference a value the source never published. The rest is specified."
  ],
  "build_warnings": null,
  "canonical": "https://prismaticalgos.com/compendium/algo/10501-news-sentiment-filter",
  "snapshot": "2026-10-31 15:46 UTC",
  "source_url": "https://www.mql5.com/en/articles/22830",
  "limitations": "These are machine-extracted research records, not verified trading systems. Catalog rank measures specification detail and claimed evidence, not profitability. The catalog extraction pipeline did not independently backtest these records. Its separate selected testing ledger does not validate the wider corpus. Rules, labels and parameters can be incomplete or wrong; consult the original source and its reuse terms."
}