/api/anomaly_accounts

Header

POST https://api.Fluxrails.app/api/v1/ai/anomaly_accounts
X-Customer-Api-Id: <uuid>
X-Secret: <secret>
Content-Type: application/json

Request Body Schema

// Request Body Schema
{
  "contamination": number,    // expected share of anomalous accounts (0–1)
  "rows": [
    {
      "id": "string",         // unique account identifier
      "features": {           // any numeric or categorical metrics
        "<key>": <number|string>,
        …
      },
      "label": 0|1|null       // optional ground-truth: 1=bad/bot, 0=good, null/unset=unlabeled
    }
  ]
}

Example Request Body

// Example Request Body
{
  "contamination": 0.05,
  "rows": [
    {
      "id": "acct-001",
      "features": {
        "age_days": 730,
        "num_logins": 120,
        "avg_txn": 45.2,
        "device_os": "ios",
        "2fa_enabled": 1
      },
      "label": 0
    },
    {
      "id": "acct-002",
      "features": {
        "age_days": 14,
        "num_logins": 75,
        "avg_txn": 3.1,
        "device_os": "android",
        "2fa_enabled": 0
      }
    },
    {
      "id": "acct-003",
      "features": {
        "age_days": 1,
        "num_logins": 350,
        "avg_txn": 0.0,
        "device_os": "windows",
        "2fa_enabled": 0
      },
      "label": 1
    },
    {
      "id": "acct-004",
      "features": {
        "age_days": 3,
        "num_logins": 280,
        "avg_txn": 999.9,
        "device_os": "linux",
        "2fa_enabled": 0
      }
    }
  ]
}

Example Response Body

// Example Response Body
{
  "kpi": {
    "detected_pct": 75.0,
    "auc": 0.6666666666666667
  },
  "details": [
    {
      "id": "acct-001",
      "anomaly_score": -0.08251618866556054,
      "fraud_flag": true
    },
    {
      "id": "acct-002",
      "anomaly_score": -0.08251618866556054,
      "fraud_flag": true
    },
    {
      "id": "acct-003",
      "anomaly_score": -0.08251618866556054,
      "fraud_flag": true
    },
    {
      "id": "acct-004",
      "anomaly_score": -0.08251618866556065,
      "fraud_flag": false
    }
  ],
  "interpretation": "75.0% of records flagged by LOF. Threshold=-0.0825 (quantile 0.95)."
}

Description

• Unsupervised anomaly detection for account-level profiles: flags the top-weird accounts based on IsolationForest, LOF, or Auto-Encoder bake-off. • `contamination` sets the expected anomaly rate and decision threshold. • Returns per-account `anomaly_score` (higher ⇒ more anomalous) and boolean `fraud_flag`. • Provides batch KPIs: `detected_pct` and optional `auc` on labeled subset. • Delivers a human-readable `interpretation` summarizing model, threshold, and flag rate.

Business Usage

• Stream your account-behavior metrics into this route for real-time risk triage. • Sort `details` by `anomaly_score` to prioritize suspicious accounts for your trust & safety team. • Monitor `detected_pct` drift to tune `contamination` or feature set. • Feed confirmed labels back on the next call—model selection becomes data-driven and AUC improves. • Automate review workflows or step-up authentication for high-score accounts.

← Back to all routes