Version history
Document type is added as a new level which allows users to manage, besides specific projects, also custom document types that can be shared among multiple projects and/or organizations.
Document types are added to the overview page
Besides creating a new document type, it is also possible to add an existing document type to a project
For an existing document type, the user can choose per project which entities to reuse and/or new entities to add
Document type settings are introduced where organizations can be managed/added who have access to the doctype
Other improvements that are part of this release:
The steps to create a project are simplified
Name of logged in user and organisation are added to the navigation bar
The number of documents in human validation and accuracy are added to the overview page
OCR speed has been increased by a factor 2x
New pipeline for training incremental models
Various performance improvements, security improvements and bug fixes
Analytics module is updated
Key metrics (%STP, time saved and number of pages processed) are added
Selection of a predefined timespan is possible
Details of status(es) are shown per doctype(s)
Graphs are made more precise with indication of amount
Last snapshot is added
Enrichments are added to Metamaze
New step in the pipeline after entity extraction and before business rules
Set-up in project settings
Name an enrichment and make it part of a composite
Webhook results can be retrieved and added to human validation
The user can enable the automatic deletion of production data after 180 days
Metamaze is able to process e-mails and its attachments
QA task module is renamed to task module
task types are introduced
type 'review' has the same functionality as before the QA task
new type 'annotation' is added, to allow for data control in the annotation process
Suggested annotations tasks are created by MetaMaze to accelerate annotations
Similar documents are grouped together
Documents are proposed that add the most value i.e. model learns from - MetaMaze Optimal Document Selection Strategy applied
Autocorrect allows to detect which documents might be misannotated
Aggregated entity values are also visible in human validation
Tooltips are added to project overview and action buttons
Automatically mark your batch as done in human validation and data annotation
Customization of view is added for tables in training and production
Get better training results based on your document type and language
choose between text or text and layout when creating or editing a document type and retrieve your choice in the document types section of project settings
choose between training one multi-lingual model or one model per language
Retraining your model is made more efficient and controlled
it's possible to send multiple relevant documents from production to training
automatically documents in human validation are taken into account
MetaMaze suggests to users QA tasks
performance of suggested tasks is tracked and shown
Design of navigation bar is updated
regrouping of modules
text of modules is always shown (instead of icons)
user can decide to open or close
Improvement of QA task module
buttons to mark as done and failed are added
parking a document is added
missing labels and added or deleted labels (in human validation) are shown
date filter is added
design is updated
Parking a document in human validation is added
when a user is uncertain about the handling of a specifc document in human validation it is possible to park it and add a reason for parking
filter on parked documents is added
Filters are (automatically) saved in human validation
if no results are found for your filter a message will be shown
MetaMaze indicates whether a filter is (still) active
Removing spaces and validation regex options are added to project settings
MetaMaze is able to filter white pages out (page management)
MetaMaze indicates which pages are not used (page management)
