Fake Candidate Checker – LinkedIn Fraud Risk Analyzer

Overview

The Fake Candidate Checker is an Apify actor designed to analyze LinkedIn profiles for potential hiring fraud risk indicators using metadata-based heuristics, explainable scoring, and optional limited public access checks.

The actor is built to remain reliable even when LinkedIn restricts automated access by falling back to non-invasive metadata-only analysis instead of failing or attempting aggressive scraping.

This makes it suitable for HR teams, recruiters, staffing agencies, and hiring managers who want early signals about profiles that may require additional verification.

---

What Problem Does This Solve?

Modern hiring processes increasingly face challenges such as:

• Fake or misrepresented candidates
• Auto-generated or incomplete LinkedIn profiles
• Inconsistent public profile signals
• Increased risk in remote hiring workflows
• Wasted interview and onboarding effort

This actor helps teams identify profiles that may require enhanced screening before progressing further in the hiring pipeline.

---

Key Features

• ✅ Metadata-only analysis (no LinkedIn login required)
• ✅ Graceful handling of LinkedIn access restrictions
• ✅ Explainable risk scores with clear flags
• ✅ Store-safe, non-invasive design
• ✅ Works reliably even when LinkedIn blocks access
• ✅ Suitable for compliance-sensitive environments

---

How It Works

1. Accepts one or more LinkedIn profile URLs
2. Optionally attempts limited public data access
3. If LinkedIn restricts access, switches to metadata-only heuristics
4. Assigns a probabilistic risk score
5. Outputs detected risk indicators and recommended next steps

---

Input

Example Input

{
  "linkedinUrls": [
    "https://www.linkedin.com/in/example-profile/"
  ],
  "metadataOnly": true
}

### Example Outputs
{
  "linkedinUrl": "https://www.linkedin.com/in/example-profile/",
  "mode": "metadata-only",
  "riskScore": 45,
  "flags": [
    "Limited public LinkedIn visibility",
    "Suspicious numeric-heavy profile handle"
  ],
  "recommendation": "Manual review recommended",
  "disclaimer": "This tool provides probabilistic risk indicators only and does not confirm fraud."
}