> For the complete documentation index, see [llms.txt](https://docs.docbits.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.docbits.com/administration-and-setup/settings/global-settings/document-types/script/scripting-in-docbits/sample-scripts/similar-document-detection.md).

# Similar Document Detection

{% hint style="info" %}
**Available from version 11.48.0** — Requires `OPENSEARCH_ENABLED` license.
{% endhint %}

## What does this script do?

Uses vector-based similarity search to find documents that are semantically similar to the current one. If a document with more than 95% similarity is found, the invoice number is flagged as potentially fraudulent or duplicate.

## Trigger

`AFTER_FORMATTING` on document type **INVOICE**

## Full Script

```python
doc_id = document_json["doc_id"]
similar = vector_search(doc_id, k=5)

for doc in similar:
    if doc["similarity_percent"] > 95:
        set_field_as_invalid(
            document_data, "invoice_id",
            f"95%+ similar to: {doc['name']} (Score: {doc['similarity_percent']}%)"
        )
        break
```

## Step-by-Step Explanation

1. **Get current document ID** from `document_json`
2. **Find similar documents** with `vector_search()` returning the 5 nearest neighbors
3. **Check similarity threshold**: If any document exceeds 95% similarity, flag it
4. **Mark as invalid** with the similar document's name and similarity score

## How Vector Search Works

Each document's OCR text is converted to a 384-dimensional vector embedding when indexed. `vector_search()` finds the nearest neighbors in this vector space using k-NN (k-Nearest Neighbors), returning documents whose content is semantically similar — even if the exact words differ.

**Use cases:**

* Fraud detection (nearly identical invoices from different "vendors")
* Duplicate detection that goes beyond exact text matching
* Finding related documents across different formats or languages

## Functions Used

* [vector\_search()](/administration-and-setup/settings/global-settings/document-types/script/scripting-in-docbits/fulltext-search-functions.md#vector_search) — Find semantically similar documents
* [set\_field\_as\_invalid()](/administration-and-setup/settings/global-settings/document-types/script/scripting-in-docbits/field-functions.md#set_field_as_invalid) — Show validation error


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.docbits.com/administration-and-setup/settings/global-settings/document-types/script/scripting-in-docbits/sample-scripts/similar-document-detection.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.