# Glossary

Our goal is to make using Alphamoon Workspace as easy as possible. We have gathered a bunch of useful terms which will make it easier to navigate our tool.

{% hint style="info" %}
**Before we dive in:** 

We are constantly improving our **Documentation**. 

If the terms you are interested in are not here yet, feel free to contact us at: **<contact@alphamoon.ai>**
{% endhint %}

### Platform-related Terms

<details>

<summary>Process</summary>

List of collections (files) representing a business process that leverages data from processed documents.

</details>

<details>

<summary>Queue</summary>

A list of collections uploaded to the selected process.

</details>

<details>

<summary>Collection</summary>

A file or list of files uploaded to the selected process.

</details>

<details>

<summary>Supervision Mode</summary>

Tool to display processed results, make corrections to the model’s errors, and export results.

</details>

{% hint style="info" %}
**Platform Walkthrough**

To give you a better idea of how our Platform is organized, go to [Dashboard](https://alphamoon.gitbook.io/documentation/~/changes/E5WHATIwrsF0u91xTwvD/platform-walkthrough/dashboard), [Queue](https://alphamoon.gitbook.io/documentation/~/changes/E5WHATIwrsF0u91xTwvD/platform-walkthrough/queue), [Configuration](https://alphamoon.gitbook.io/documentation/~/changes/E5WHATIwrsF0u91xTwvD/platform-walkthrough/configuration), and [Supervision Mode](https://alphamoon.gitbook.io/documentation/~/changes/E5WHATIwrsF0u91xTwvD/platform-walkthrough/supervision).
{% endhint %}

### Tech-related terms

Interested in how our technology works? Take a look at these terms:

<details>

<summary>Intelligent Document Processing (IDP)</summary>

Intelligent Document Processing (IDP) is a technology-driven approach that combines artificial intelligence (AI), machine learning (ML), and natural language processing (NLP) to automate the extraction, understanding, and processing of information from various types of documents. It involves the use of advanced algorithms and techniques to analyze and extract data from unstructured documents, such as invoices, forms, contracts, receipts, and emails

</details>

<details>

<summary>Optical Character Recognition (OCR)</summary>

Optical Character Recognition is the functionality in which the document is scanned by the platform and “translated” from image to text. This means that the tool converts all of the contents into a machine-encrypted text

</details>

<details>

<summary>Data Extraction</summary>

For extraction to happen, Optical Character Recognition (OCR) needs to happen first. Named Entity Recognition (NER) is a natural language processing (NLP) technique used to identify and classify named entities within text. Named entities are specific words or phrases that refer to real-world entities such as persons, organizations, locations, dates, time expressions, quantities, monetary values, and more.&#x20;

Once the image is converted, the platform pulls the relevant data. Relevant means user-defined or pre-defined in the platform. The data is then downloadable as a CSV, XLSX, or JSON, depending on your needs.

</details>

<details>

<summary>Natural Language Processing (NLP)</summary>

Natural Language Processing (NLP) is a subfield of artificial intelligence (AI) that focuses on the interaction between computers and human language. It involves the use of algorithms and techniques to enable computers to understand, interpret, and generate natural language in a way that is meaningful and useful.

</details>

<details>

<summary>Zero-Shot Model</summary>

The Zero-Shot Model removes data annotation from the equation therefore the model can process any document.

</details>

#### Upcoming Feature

<details>

<summary>Continual Learning</summary>

Continual learning refers to the ability of an AI system to continuously acquire and integrate new knowledge and skills over time while retaining and building upon previously learned information. Unlike traditional machine learning approaches that require training models from scratch with new data, continual learning aims to enable AI systems to learn incrementally and adapt to evolving environments without catastrophic forgetting.

</details>

{% hint style="info" %}
**Tip:**&#x20;

Want to stay up-to-date with our tech content? Feel free to visit our [Blog](https://alphamoon.ai/category/blog/).
{% endhint %}

### Roles & Permissions

You can access Alphamoon Workspace either as an **Admin** or as a **User**.&#x20;

| Role      | Permissions                                                                                                                                                                                                                                                                                               |
| --------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Admin** | <ul><li>Has access to the <strong>User Management</strong>, <strong>Your profile</strong>, and <strong>Documentation</strong> tabs.</li><li>Can access, create, rename, and delete processes</li><li>Can upload files and folders</li><li>Is the <strong>process owner</strong> of each process</li></ul> |
| **User**  | <ul><li>Has access to <strong>Your profile</strong>, and <strong>Documentation</strong> tabs </li><li>Can access processes, but not create new ones </li><li>Can upload files and folders</li></ul>                                                                                                       |

You can also manage **User Access** rights within a selected process:<br>

| Role              | Permissions                                                                                                                                    |
| ----------------- | ---------------------------------------------------------------------------------------------------------------------------------------------- |
| **Process Owner** | <ul><li>Has a <strong>full control</strong> over a given process.</li></ul>                                                                    |
| **Editor**        | <ul><li>Can <strong>upload</strong> and <strong>access files</strong> in a process.</li><li>Can <strong>edit</strong> the results.</li></ul>   |
| **Viewer**        | <ul><li>Can <strong>access files</strong> uploaded to a process.</li><li>Can not upload new files.</li><li>Can not edit the results.</li></ul> |

### Statuses

During processing, the collection will be in one of the following statuses:

<table><thead><tr><th width="184">Status</th><th>Description</th></tr></thead><tbody><tr><td>Waiting for files</td><td>A new collection has been created but all the files have not been uploaded yet.</td></tr><tr><td>Queued</td><td>Your collection is prepared and its files are automatically queued for processing.</td></tr><tr><td>In progress</td><td>Your collection is being processed by Alphamoon Workspace.</td></tr><tr><td><a data-footnote-ref href="#user-content-fn-1"><mark style="color:orange;">Needs review</mark></a></td><td>Your collection has been processed, but it did not meet the certainty threshold for detected fields and requires human supervision as a result.<br><br>You will also see this status when the model has not found the fields marked as <strong>Mandatory</strong>.</td></tr><tr><td><mark style="color:green;">Accepted</mark></td><td>The files in your collection have been successfully processed and listed on the Queue page.</td></tr><tr><td><mark style="color:red;">Failed</mark></td><td>Your collection or file has been corrupted during the upload or processing stage. Try to delete the collection and re-upload it.<br><br>When you hover over the name of the status, information will appear on what might have gone wrong.</td></tr></tbody></table>

{% hint style="success" %}
**Awesome!**

Looks like you already know all the necessary terms to easily navigate through our Platform.
{% endhint %}

####

#### See next:

[^1]: