DoxAI Documentation & Help Center

Learn how to build and manage powerful AI-driven digital solutions. Get documentation, example code, tutorials, and more.

QUICK ACCESS

What is Categorise and Split AI?

Categorise and Split AI is an intelligent document categorisation and splitting solution that leverages advanced artificial intelligence and machine learning algorithms to automatically classify, organise and split your documents with unparalleled accuracy and efficiency.

What is the difference between Categorise AI and Split AI?

Categorise AI is an AI automated document classification tool that detects and labels the main and subcategories of each document or section. Split AI, on the other hand, enables automated PDF splitting based on those detected categories. Businesses looking to organise documents using AI can choose either classification alone or a combined solution for full automation.

How does Categorise AI work?

Categorise AI uses advanced AI and machine learning algorithms to analyse the content of documents and then assign them to specific categories, automating the categorisation process.

What are the key features of Categorise AI?

The key features are as follows:

  • Automated Categorisation of data and documents
  • Custom AI Categories
  • Over 50 Document Categories
  • More than 40 Dataset Types
  • Automated document conversion
  • 17+ Templates

What problem does Categorise AI solve?

Categorise AI solves the issue of manually classifying documents and data, saving time, and reducing errors in the classification process. The main advantage of Categorise AI is its ability to categorise large volumes of documents, boosting operational efficiency and reducing errors quickly and accurately.

Can Categorise AI be used for filing, archiving, or indexing?

Definitely. Categorise AI is widely used as an AI document filing and archiving solution, especially for companies migrating to digital or knowledge-based systems. It tags, splits, and indexes documents making them easy to retrieve and manage in systems like SharePoint, CRMs, or content libraries and more.

Can Categorise AI identify multiple document types within the same file?

Absolutely. Categorise AI uses artificial intelligence to split PDFs by content, even when multiple categories exist on the same page. It is designed for complex document sets like financial statements or legal packets, helping you automatically classify and separate documents for streamlined processing.

How does Categorise AI optimise business processes?

Categorise AI optimises business processes by swiftly and accurately processing a large volume of documents, reducing the need for extensive human intervention, and minimising errors.

Can Categorise AI be customised for our business workflows?

Yes. Through custom AI document categories, you can train the model to classify based on your unique needs such as sorting payslips by employee, or invoices by vendor. This makes it ideal for businesses seeking AI automation tailored to specific industry workflows, such as legal, finance, insurance, professional or healthcare and more.

What industries benefit the most from Categorise AI?

Industries with high document volumes and compliance requirements benefit the most. These include:
Banking: Loan document automation, KYC, and compliance
Legal: Case file organisation, contract classification
Healthcare: Medical record filing and archiving
Accounting: Automated tax document and receipt categorisation

If you are searching for AI document automation for finance, legal firms, accounting firms or healthcare this solution is a strong fit.

What kind of cost reduction can be achieved with Categorise AI?

Categorise AI can reduce costs by up to 98% by streamlining document categorisation processes and eliminating the need for significant manual interventions.

How fast is Categorise AI compared to manual processing?

Categorise AI is 100x faster than manual classification, enabling real-time processing of high-volume documents

Does Categorise AI adapt and improve its accuracy over time?

Categorise AI employs machine learning, which means it can learn from its interactions and improve its accuracy over time as it processes more documents.

Is Categorise AI secure to use?

Yes, Categorise AI is SOC2 Type 2 certified and adheres to compliance and regulatory standards, ensuring a secure environment for document categorisation.

What is the format of the Categorise AI output?

Categorise AI returns output in structured JSON format with URLs for downloading the split files. It includes document IDs, main categories, subcategories, and page ranges. This is ideal for teams wanting integrated AI document workflow solutions that connect with downstream systems.

Is there a feedback or learning loop in Categorise AI?

Yes. A built-in feedback mechanism allows users to correct misclassifications. Over time, the machine learning-based document categorisation model improves, making it more precise for your unique use cases, especially useful for custom AI categorisation in regulated industries.

Can Categorise AI handle diverse types of datasets and predefined document categories?

Yes, Categorise AI supports more than 40 dataset types and provides over 50 predefined document categories, ensuring versatility in handling several types of datasets and document categories.

What types of documents can Categorise AI handle?

Categorise AI can handle a wide range of document types, including text documents, PDFs (PORTABLE DOCUMENT FORMAT), spreadsheets, presentations, structured, unstructured documents and more.

Can I train Categorise AI to understand specific business needs related to my industry?

Yes, Categorise AI offers custom AI categories, which means you can train the system to recognise industry-specific documents, content, and structures as per the business need.

What kind of customer support does Categorise AI offer?

We provide comprehensive customer support to assist you with any inquiries or issues you may encounter. Our support team is available via email and phone to ensure a smooth experience. You can also reach out to us on: Contact us.