This document provides a comprehensive overview of using the Copyleaks API to detect text manipulation attempts in submitted documents. Text manipulation detection helps identify when users attempt to deceive detection systems through various deceptive techniques.

Introduction

The Text Manipulation detection feature is designed to identify sophisticated attempts to bypass detection systems. This feature recognizes when users employ various deceptive techniques to hide copied content or manipulate the scanning process.

Text manipulation attempts can include:

• Hidden Characters: Inserting invisible characters to break up text patterns
• Character Replacement: Using special characters or symbols that look similar to normal letters
• Invisible or White Text: Adding white text on white backgrounds or other concealment methods (works only in PDF and DOCX documents)
• Major Text Exclusion: Attempting to exclude large portions of text from scanning

By detecting these manipulation attempts, you can maintain the integrity of your plagiarism detection or AI detection process and ensure accurate results.

Before You Begin

To get the most out of this document, you should first be familiar with how to submit a basic scan. If you’re new to the process, we recommend starting with the guide below.

Note

Detect Plagiarism in Text: This guide walks you through the fundamentals of customizing your API request to scan for plagiarism.

Getting Started

Enabling Text Manipulation Detection

To enable text manipulation detection in your scans, set the properties.cheatDetection parameter to true:

{
  "properties": {
    "cheatDetection": true
  }
}

Default: false

When enabled, the submitted document will be analyzed for various text manipulation techniques. If manipulation is detected, a scan alert will be added to the completed webhook.

For more information on submitting documents, check out our documentation for URL, OCR, and File scans.

Interpreting Scan Results

Scan Alerts

When text manipulation is detected, you’ll receive specific alerts in your scan completion webhook. These alerts are found at:

notifications.alerts[]

Types of Text Manipulation Alerts

Alert Code	Title	Description
suspected-cheating-detected	Advanced Detection: Hidden Characters	Detected possible use of hidden characters to cheat the plagiarism scan
suspected-character-replacement	Advanced Detection: Character Replacement	Detected possible use of special characters to cheat the plagiarism scan
suspected-white-text	Suspected Cheating: Invisible Text	Detected possible use of invisible or white text - switch to textual version to see all text
text-mostly-excluded	Advanced Detection: Major Text Exclusion	Detected possible attempt to exclude the majority of text from scanning
cheat-detection-failed	Advanced Detection Failed	Unable to validate that there was no manipulation in the submitted document

For a complete list of all possible alerts, see our Scan Alerts documentation.

Example Alert Response

{
  "notifications": {
    "alerts": [
      {
        "code": "suspected-character-replacement",
        "title": "Advanced Detection: Character Replacement",
        "message": "We have detected possible use of special characters to cheat the plagiarism scan.",
        "category": 3
      }
    ]
  }
}

Best Practices

• Monitor alerts: Check for text manipulation alerts in your webhook responses
• Document findings: Keep records of detected manipulation attempts for policy enforcement
• Handle failures: Implement proper error handling for cases where detection fails
• Stay updated: Alert titles and messages may change over time as the system improves

Next Steps

After implementing text manipulation detection:

Detect Plagiarism in Text Explore additional scanning capabilities beyond text manipulation detection.

How to Display Scan Reports Learn how to present detection results to users effectively.

Support

Should you require any assistance or have inquiries about implementing text manipulation detection, please contact Copyleaks Support or ask a question on StackOverflow with the copyleaks-api tag. We appreciate your interest in Copyleaks and look forward to supporting your efforts to maintain originality and integrity.

By implementing text manipulation detection, you’re adding an essential layer of security to your plagiarism detection workflow, ensuring that sophisticated cheating attempts don’t go unnoticed.

Need advanced content analysis?

Get personalized guidance on implementing comprehensive content analysis with text manipulation detection and other advanced features.

Book a Demo