Scan Content for Plagiarism

The "Scan Content for Plagiarism" step allows you to check text for potential plagiarism by comparing it against billions of web pages, academic papers, and other published sources. This powerful tool uses Originality.ai's advanced plagiarism detection technology to help you ensure content originality and avoid copyright issues.

How to Configure the Scan Content for Plagiarism Step

Configuring the step requires setting the parameters shown below:

[IMAGE: Screenshot of the Scan Content for Plagiarism step configuration panel]

Text

Enter the text content you want to scan for plagiarism. This can be a paragraph, article, or any written content you need to verify for originality.

You can use Liquid syntax to dynamically reference text from a previous step:

{{ step_1.output.content_to_check }}

Minimum Length (Optional)

The minimum text length required for reliable plagiarism detection. If your text is shorter than this value, the step will return a warning about potential reduced accuracy.

Default: 100 characters

Match Sensitivity (Optional)

Adjust how sensitive the plagiarism detection should be. Higher sensitivity will detect smaller matches and paraphrased content, while lower sensitivity will only flag more substantial, direct matches.

Range: 1-5 (1 = low sensitivity, 5 = high sensitivity) Default: 3

Output Format

The Scan Content for Plagiarism step returns a JSON object containing detailed analysis results:

{
  "original_text": "The first few sentences of the analyzed text...",
  "originality_score": 0.85,
  "is_original": true,
  "plagiarism_score": 0.15,
  "analysis": {
    "originality_percentage": 85,
    "plagiarism_percentage": 15,
    "matches": [
      {
        "matched_text": "This specific section of the content appears to match an existing source.",
        "similarity": 0.92,
        "source_url": "https://example.com/original-article",
        "source_title": "Original Article Title",
        "source_date": "2023-08-15"
      },
      {
        "matched_text": "Another section that shows similarity to published content.",
        "similarity": 0.88,
        "source_url": "https://anothersource.com/blog-post",
        "source_title": "Related Blog Post",
        "source_date": "2024-01-10"
      }
    ]
  },
  "word_count": 325,
  "character_count": 1842,
  "language": "en"
}

[IMAGE: Screenshot showing an example of the Scan Content for Plagiarism results in the AirOps interface]

Understanding the Output Metrics

• original_text: The first portion of the analyzed text (truncated for display purposes)

• originality_score: A score from 0 to 1 indicating the overall originality of the content (higher = more original)

• is_original: A boolean value indicating if the system considers the content adequately original (true/false)

• plagiarism_score: A score from 0 to 1 indicating the overall level of potential plagiarism (higher = more potential plagiarism)

• analysis: Detailed information about the plagiarism detection

  • originality_percentage: Originality score expressed as a percentage

  • plagiarism_percentage: Plagiarism score expressed as a percentage

  • matches: Array of specific matched text segments found during the analysis

    • matched_text: The text segment that matches an existing source

    • similarity: How similar the match is to the source (0-1 scale)

    • source_url: The URL of the matching source

    • source_title: The title of the matching source

    • source_date: The publication date of the matching source

• word_count: Total number of words in the analyzed text

• character_count: Total number of characters in the analyzed text

• language: Detected language of the content

How to Use the Results

You can access the plagiarism scan results in subsequent workflow steps using Liquid syntax. For example:

To access the originality score:

{{ step_2.output.originality_score }}

To check if content is considered original:

{{ step_2.output.is_original }}

To loop through matched sources in an LLM prompt:

{% for match in step_2.output.analysis.matches %}
- Matched text: "{{ match.matched_text }}"
  - Source: {{ match.source_title }}
  - URL: {{ match.source_url }}
  - Similarity: {{ match.similarity | times: 100 }}%
{% endfor %}

Common Use Cases

• Verifying the originality of submitted content

• Quality control for content creation workflows

• Checking AI-generated content for potential plagiarism

• Ensuring academic integrity of papers and articles

• Pre-publication review to avoid copyright issues

• Validating guest blog submissions

• Reviewing website content before launch

How to Continue if the Scan Fails

By default, the Scan Content for Plagiarism step will terminate the workflow if it fails. To continue the workflow if the step fails, click "Continue" at the bottom of the settings panel.

[IMAGE: Screenshot showing the "Continue" option in the step settings]

The step will return the following keys when it fails:

• output: this will be null

• error:

  • message: the message returned from the step

  • code: the error code representing the error

Common error causes include:

• Text too short for reliable analysis

• Unsupported language

• Service unavailability

Example Workflow

Here's a common pattern using the Scan Content for Plagiarism step:

1. Start with content submitted for publication or review

2. Use the "Scan Content for Plagiarism" step to analyze the content for originality

3. Use a Conditional step to branch the workflow based on the plagiarism score:

   • If plagiarism score is high (>0.3): Reject and return for revision

   • If plagiarism score is moderate (0.1-0.3): Route to human review

   • If plagiarism score is low (<0.1): Proceed with normal publication

4. For content requiring revision or review, use an LLM step to:

   • Identify specific problematic sections

   • Suggest potential rewrites

   • Generate a detailed feedback report with sources of potential plagiarism

5. Output the appropriate response based on the plagiarism analysis

This workflow enables you to implement a robust content originality policy, catching potential copyright issues before publication while providing constructive feedback for content improvement.