Recent Posts

Recent Video

 

Pattern Search using Artificial Intelligence

Pattern Search using Artificial Intelligence

Leveraging the power of artificial intelligence and machine learning, Docxonomy gives the ability to predefine patterns that you want to look for and find in your files. This ability is an incredibly powerful feature that can be leveraged in many circumstances like GDPR, regulatory, legal e-discovery as well as many others areas of your business.

 

Imagine trying to find all documents in your system that contain a Social Security Number (SSN) – not a specific number – but any social security number. In a traditional search system, that is not possible. You can’t just tell it to go off and find them all. Docxonomy comes with this ability built-in. Additionally, the patterns our system uses are customizable, and you can create your own. Say, you need to find where all the model numbers of the widgets you manufacture are found within your product documentation. Directly apply these patterns to your account, and we’ll find those files, regardless of where they are located, as well as any new or updated files that are indexed by the system. We call this feature “Intelligent Discovered Metadata.”

 

When the pattern is identified, the system automatically flags the file as having that particular pattern within it. Moreover, when flagged, we can optionally trigger a whole host of actions that can be taken on the backend. For example, whenever a sensitive piece of information is found (again, say an SSN), we can send a notification that action should be taken. Finally, once it’s configured, you don’t have to do anything else, the system will monitor it for you.

 

That sounds easy you might say when you are dealing with text-based files like Word or PowerPoint or Excel but what happens if you have an image? What happens if the pattern is spoken or displayed on a slide in a video? These are issues that make the process of automated intelligence a whole lot harder when trying to identify patterns in files. Docxonomy is able to detect patterns in such videos and images and transcribed speech also!

 

You may be thinking, “But I have thousands (or millions!) of files in other repositories. I can’t afford to move them to Docxonomy”. Well, you don’t have to! With our powerful connector technology, we can index data into Docxonomy, allowing our intelligent search to do its work, even against files that are stored in on-premise or other cloud systems.

 

Let’s say for example you have a Documentum server, several file servers, and a Box™ cloud repository. All of these solutions are important and are running your mission-critical business processes. Obviously, all of these processes have taken a long time to develop and build, and there are many users within your organization using them on a daily basis. With Docxonomy connectors, you can leave all the files where they are and index them automatically. We don’t force you to migrate all that data unless you want to. However, now that Docxonomy has indexed the files we can leverage “Intelligent Discovered Metadata” to identify the patterns described above.

 

It’s an incredibly powerful feature of Docxonomy, and we are excited to share it with all of our users. If you would like to learn more about this feature or sign up to be a user, click the link below to get started for free.

 

Try It Contact Us
Bryan Reynolds

Bryan Reynolds has over 25 years of experience as a successful entrepreneur, senior executive and managing consultant with core competencies focused on enterprise content management, mobility, business process engineering, imaging, and records management. Currently, Mr. Reynolds is the Founder and CEO of Docxonomy. The breadth of his knowledge includes the architectural design and development as well as project management of numerous global, large-scale document/records management initiatives across multiple industries including pharmaceutical, biotechnology, medical devices, financial services, insurance, healthcare and public sector.

No Comments

Post a Comment

Comment
Name
Email
Website