3. Learn. Connect. Collaborate.
About GDPR
The GDPR regulation applies to organizations which collect and process personal
data. It aims to give more control to individuals over usage of their personal data.
Privacy is a growing concern in our connected, digital world. We see increasing
challenges related to securing our digital footprint.
4. Learn. Connect. Collaborate.
GDPR Problem
Right to forget - Under the new GDPR, organizations around the world must not
only protect personal data but also build mechanisms to forget personal data on
request from individuals.
The GDPR Watchdog shows our approach to solve this problem by integrating
computer vision and Natural language processing with the Alfresco Repository.
.
5. Learn. Connect. Collaborate.
Computer
Vision
&
Machine
Learning
We Humans use our eyes and brains to gather
information and make decisions about the world
around us.
Computer Vision and Machine Learning aim to give
a similar, if not better, capability to a machine or
computer.
7. Learn. Connect. Collaborate.
TML Sub-system • Specific functional task, to provide node
inception services using machine
learning.
• Isolated Spring application context
• Can be started | stopped independent of
the repository
• Can be configured in runtime via JMX
• ChildApplicationContextFactory
– Category : TML (Texter Machine Learning)
– Type : Gdpr
8.
9. Learn. Connect. Collaborate.
Tesseract Integration – Custom transformers to OCR
We are executing OCR at the
repository side using Tesseract
to execute Text extraction from
images or documents.
The GDPR watchdog processes
both images (using computer
vision) and text (using Natural
language processing)
10. The GDPR service webscript takes a document reference as unique parameter
and triggers a series of node inception operations to detect GDPR data.
GDPR Watchdog – Gdpr Webcript Node Inception Service
11. GDPR Watchdog – Repository Action
This repository action allows to traverse the full repository, with mime-type filters.
This action execution can be scheduled with asynchronous execution
14. Learn. Connect. Collaborate.
TML Gateway is designed to be open for new integrations via its set of pluggable
interfaces. At any time we can plug-in a new interface that can be executed sequentially
to enrich the output data. We can also train a specific model to adjust to specific
customer necessities.
Adding new ML tools and logic will be transparent and will not need any service
downtimes.
15. Learn. Connect. Collaborate.
Why TML Gateway?
• Can integrate with several engines
• New ML engines can be added via
pluggable interfaces
• Gateway itself can be easily
customized to meet new requirements
– Call ML engines based on a custom
logic (even using machine learning)
– Engines can be called in parallel or
can live on a hybrid-system
• Simple REST API, can be used on
several other systems
– Why not integrate on a APS
workflow?
16. Learn. Connect. Collaborate.
TML Gateway - API
• Input
– File to be recognized
• Output
– Region of Interest: area where sensitive data was detected
– Information type: can be Face, Name, Address, License plate, Email,
VAT number, Phone number
– Associated information: the recognized data
– Trust: confidence level of the detected data
17.
18.
19. The Right to Forget
GDPR WatchDog – Content Anonimization
27. Learn. Connect. Collaborate.
Computer
Vision
&
Machine
Learning
We’ve seen an integrated set of technologies that
provide automatic extraction, analysis and
understanding of useful information from a single
piece of content or a set of related contents.
These can be from several different sources:
images, videos, text, documents, spreadsheets,…
28. Learn. Connect. Collaborate.
Advantages of TML over public Cloud ML Services
ü Optimized for customer scenario
ü Maximized results and automation
ü Ability to use several different engines
simultaneously
ü Guarantees privacy of data and models
ü Seamless integration
Texter Machine Learning
29. Learn. Connect. Collaborate.
With TML you have a turnkey AI
solution delivered to you that:
• Automates operations
• Dramatically increases
efficiency
• Processes and analyses data
in new ways beyond human
capabilities
• Is a cornerstone of the Digital
Transformation path
TexterBlue Machine Learning