first commit

backend/app/rules/assessment_rules.py

@@ -0,0 +1,57 @@
+"""Rule-based fraud amount assessment.
+
+Classifies each transaction into high / medium / low confidence fraud,
+and generates initial reason text.
+"""
+from app.models.transaction import TransactionRecord
+from app.models.assessment import ConfidenceLevel
+
+FRAUD_KEYWORDS = ["投资", "保证金", "手续费", "解冻", "税费", "充值", "提币", "提现"]
+
+
+def classify_transaction(tx: TransactionRecord) -> tuple[ConfidenceLevel, str, str]:
+    """Return (confidence_level, reason, exclude_reason)."""
+    if tx.is_transit:
+        return (
+            ConfidenceLevel.low,
+            f"该笔为本人账户间中转（{tx.source_app.value} -> {tx.counterparty_name}），不直接计入被骗损失。",
+            "本人账户间互转，仅作为资金路径展示。",
+        )
+
+    if tx.direction.value == "in":
+        return (
+            ConfidenceLevel.low,
+            f"该笔为收入方向交易（+¥{float(tx.amount):,.2f}），通常不属于被骗损失。",
+            "收入交易不计入损失。",
+        )
+
+    remark = tx.remark or ""
+    counterparty = tx.counterparty_name or ""
+    confidence = tx.confidence
+
+    has_fraud_keyword = any(kw in remark or kw in counterparty for kw in FRAUD_KEYWORDS)
+
+    if confidence >= 0.9 and has_fraud_keyword:
+        return (
+            ConfidenceLevel.high,
+            f"受害人经{tx.source_app.value}向「{counterparty}」转账¥{float(tx.amount):,.2f}，"
+            f"备注为「{remark}」，与诈骗常见话术吻合，OCR置信度{confidence:.0%}。",
+            "",
+        )
+
+    if confidence >= 0.85:
+        reason = (
+            f"受害人经{tx.source_app.value}向「{counterparty}」转账¥{float(tx.amount):,.2f}。"
+        )
+        if has_fraud_keyword:
+            reason += f"备注「{remark}」含涉诈关键词。"
+            return ConfidenceLevel.high, reason, ""
+        reason += "建议结合笔录确认是否受诱导操作。"
+        return ConfidenceLevel.medium, reason, "如经核实该笔为受害人主动日常消费，应排除。"
+
+    return (
+        ConfidenceLevel.medium,
+        f"受害人经{tx.source_app.value}向「{counterparty}」转账¥{float(tx.amount):,.2f}，"
+        f"OCR置信度较低({confidence:.0%})，需人工复核。",
+        "OCR置信度不足，可能存在识别误差。",
+    )

backend/app/rules/dedup_rules.py

@@ -0,0 +1,32 @@
+"""Transaction deduplication rules.
+
+Determines whether two transaction records likely represent the same
+underlying financial event captured from different screenshots / pages.
+"""
+from datetime import timedelta
+
+from app.models.transaction import TransactionRecord
+
+TIME_WINDOW = timedelta(minutes=5)
+
+
+def is_duplicate_pair(a: TransactionRecord, b: TransactionRecord) -> bool:
+    # Rule 1: exact order_no match
+    if a.order_no and b.order_no and a.order_no == b.order_no:
+        return True
+
+    # Rule 2: same amount + close time + same account tail
+    if (
+        float(a.amount) == float(b.amount)
+        and a.trade_time
+        and b.trade_time
+        and abs(a.trade_time - b.trade_time) <= TIME_WINDOW
+    ):
+        if a.self_account_tail_no and b.self_account_tail_no:
+            if a.self_account_tail_no == b.self_account_tail_no:
+                return True
+        # same counterparty and close time is also strong signal
+        if a.counterparty_name and a.counterparty_name == b.counterparty_name:
+            return True
+
+    return False

backend/app/rules/transit_rules.py

@@ -0,0 +1,35 @@
+"""Transit (self-transfer) detection rules.
+
+Identifies transactions that are internal transfers between the victim's
+own accounts (e.g. bank -> Alipay -> WeChat) and should NOT be counted
+as fraud loss.
+"""
+from app.models.transaction import TransactionRecord
+
+SELF_KEYWORDS = ["本人", "自己", "余额", "充值", "提现", "银行卡转入", "银行卡充值"]
+
+
+def is_self_transfer(tx: TransactionRecord, known_self_accounts: list[str]) -> bool:
+    """Check if a transaction is an inter-account transfer by the victim."""
+    counterparty = (tx.counterparty_name or "").lower()
+    remark = (tx.remark or "").lower()
+
+    # Rule 1: counterparty matches known self accounts
+    for acct in known_self_accounts:
+        if acct and acct.lower() in counterparty:
+            return True
+
+    # Rule 2: counterparty contains self-transfer keywords
+    for kw in SELF_KEYWORDS:
+        if kw in counterparty or kw in remark:
+            return True
+
+    # Rule 3: counterparty references another payment app owned by victim
+    app_keywords = ["支付宝", "微信", "银行卡", "数字钱包"]
+    victim_patterns = [f"{app}-" for app in app_keywords] + app_keywords
+    for pat in victim_patterns:
+        if pat in counterparty:
+            if tx.direction.value == "out":
+                return True
+
+    return False