Multi-lingual skill mapping using our default taxonomy is now fully supported across all 50+ languages when using our NextGen Resume Parser.

What’s New?

While we’ve always extracted raw skills from resumes in all supported languages, our default skills mapping that utilises the Lightcast skills taxonomy has been expanded to match skills across all common languages.

Benefits for Customers

• Standardized Data: Regardless of the language of the resume, skills are mapped to a common taxonomy, ensuring consistent data across your applications.
• Improved Usability: Mapped skills are ready for analytics, reporting, and downstream processes, eliminating the need for manual data normalization.

Key Details

• Enabled by Default: This update is automatically live for all existing and new Collections using our NextGen Resume Parser with the Lightcast skills taxonomy enabled.
• Default Output in English: By default, mapped skills are returned in English. However, if you’d like them returned in the language of the original resume, we can enable this for your account.

If you’d like to customize this setting or learn more, don’t hesitate to reach out to our team.

We’re excited to share that our invoice processing models now feature instant learning from a single document. This upgrade uses the latest advancements in AI technology to deliver greater automation benefits to customers.

What’s New?

Learn from Just One Document:
Our system now only needs one example of a particular invoice format to perform accurate data extraction. This eliminates the need to upload multiple examples for model training.

Immediate Adaptability:
Your time is valuable. When you upload a single invoice of a specific format, our models will immediately apply the learnings to the next document of the same type. This minimizes the setup and preparation time.

Higher Accuracy:
With continuous improvements driven by state-of-the-art AI, this latest release excels at recognizing layouts and specific data structures. This ensures high accuracy without extensive training.

Benefits:

• Efficiency: Reduces the effort needed to achieve accurate data extraction with just one example document.

• Speed: Models adapt instantly to new documents, streamlining your workflow.

• Precision: The latest AI advancements lead to better recognition of complex fields and layouts, improving overall accuracy.

We're excited to announce that Affinda has launched an upgraded version of our document splitting model, delivering enhanced performance for Accounts Payable document types.

Customers can easily activate this feature by selecting the new model within their Workspace settings. With this improvement, when a file is uploaded, Affinda will automatically detect and split multiple documents (such as various invoices) within the original file. This enhancement not only saves you time but also ensures that no documents are overlooked.

The upgraded Accounts Payable splitting model is now available at no cost to all our customers. It comes pre-configured to automatically split documents whenever a new invoice or other AP document is identified within a file.

For those with specific requirements, we also offer the ability to implement custom splitting models tailored to your unique use cases and document processing needs. If you’re interested in creating a custom splitting model, please contact Affinda to learn more.

We've updated the auto-validation feature in our app, enhancing how our ability to delivery straight-through processing of your documents to reduce the level of user intervention.

What's New?
You now have three customizable options for auto-validation:

1. Never: Auto-validation is disabled.
2. Validation Rules Passed: Auto-validates if all validation rules applied to that field are satisfied. In this case, field confidence will have no bearing on whether the field is auto-validated or not.
3. Validation Rules + Confidence Threshold Met: Both the confidence threshold and all validation rules must be satisfied for auto-validation. You can set the desired confidence threshold directly in the app.

These settings can be applied both at the Collection level and at a Field level within the Collections Settings page. Existing Collections remain unaffected while new Collections will have a confidence threshold not applied as default.

For more information on Validation Rules, see Machine Validation Engine.

The Affinda solution applies a range of post-processing steps to the raw text extracted from a document to ensure it is usable for downstream processing. Examples of this include:

• Formatting dates into the standard ISO format
• Removing any symbols or characters from currency values
• Matching the raw text in the document to a defined list of allowed values

To provide greater control and visibility to customers, we have introduced a new setting that allows users to view the 'raw' text extracted from the document alongside the 'parsed' value that has been formatted and post-processed. This means that users can see both values side by side and ensure that the post-processing is accurate.

How to enable this setting?

The setting can be enabled in the Validation Tool by clicking the '...' and selecting 'Show Raw Values'. It can also be disabled in the same way by clicking 'Hide Raw Values'. This setting will persist across user settings.

To meet more customer's requirements, we have released a new option that enables customers to parse documents materially faster. The option is aimed at customers who need real-time responses where seconds count. This option is particularly relevant for Resume Parsing customers using our NextGen Resume Parser.

What's New?

We've introduced an advanced feature that allows faster processing of documents by bypassing certain steps that are not always necessary. To benefit from this new feature, simply adjust the document submission parameters as follows:

• enableValidationTool: False
• deleteAfterParse: True
• compact: True

How Does It Work?

By setting these parameters, Affinda can bypass the need to save any data to our database, which eliminates unnecessary processing time and reduces the overall time taken to return results. However, note that this means that:

• The document can not be viewed in out Validation Tool (e.g. for 'human in the loop' validation)
• The document is not retained in our system so responses can not be fetched at a later date
• Field metadata is not returned, only the 'parsed' value

Customers can now sort by Collection within the Workspace view in the Affinda web app. This gives customers who are using our document classification capabilities the ability to quickly access all documents that have been left 'Unclassified' as a suitable Collection for that document has not been identified.

To align with the other documents offered by Affinda, the Job Description Parser is now visible in the Affinda Document UI through our web app. This means customers can now assess the data that has been extracted from a job description quickly and easily, without having to wade through JSON or XML formats.

Note, not all data will be visible in the Document UI (e.g. Occupation Classification) so customers should also inspect the API response to understand all that is available from the Job Description Parser.

Introducing our NextGen Resume Parser - the most accurate parser on the planet.

What’s New?

• Unmatched accuracy – the NextGen Resume Parser outperforms competitors in accuracy across key fields, by up to 50%.
• Extensive data coverage – Discover an AI parser that effortlessly extracts 100+ fields and supports over 56 languages.
• Superb adaptability - the NextGen Resume Parser can handle unconventional resume formats and industry-specific jargon, easily adapting to any business or industry.

What are the Benefits?

• Experience a seamless hiring experience - Extract the most accurate data from resumes to make fast and highly informed decisions about candidates
• Simplify screening, shortlisting and placement tasks - Access a customisable solution that can understand the fields and data you need
• Optimise your recruitment processes – Meet the Resume Parser that adapts to your unique business needs, enhancing overall candidate and recruiter satisfaction

Getting Started

If you're an existing customer, we've added 200 free credits to your account so you can start using the NextGen Resume Parser right now. Simply select “Resumes NextGen (V4)” under “Create a Workspace and Collection” within the Affinda web app.

For new customers, simply sign up for a free trial and follow the above instructions.

Find out more on how to get started with the NextGen Resume Parser here.

We have released our new Machine Validation Engine to give customers an even greater ability to achieve automation for their documents. Within the Affinda platform, rules can be configured to validate automatically if the data extracted by our AI models meets certain business logic, and therefore does not need any user intervention or review. Ultimately, this means that straight-through processing of documents can be acheived to remove the 'human in the loop' from the process. The rules can also be used to flag exceptions where the data does not meet expected values, ensuring accurate data for downstream processing.

Types of operations

Some of the standard operations that the Machine Validation Engine supports are:

• Regex - does the data extracted meet a defined regex pattern (e.g. IBAN number = [a-zA-Z]{2}[0-9]{2}[a-zA-Z0-9]{4}[0-9]{7}([a-zA-Z0-9]?){0,16})
• Equality Operation - does one field equal another field or value (e.g. does invoice total = invoice due or is there some other element like a partial payment)
• Greater Than / Less Than - check if one field is >/< another or value (e.g. due date > invoice date, invoice date < today)
• Sum - check if the sum of multiple fields / values equals another field / value (e.g. line item totals = document total, subtotal + tax = total)
• Length - check that number of characters in field is within specified range (e.g. is PO number 8 digits?)
• ABN Validation - confirms that ABN is valid

If you have other rules you would like to setup, please speak to the Affinda team.

How to enable validation rules?

The Machine Validation Engine is a paid add-on. If you would like to configure validation rules, please speak to the Affinda team who can discuss the options available in more detail and what would work best for your use case.