Information management has changed from pure document management and archiving into a real business enabler. Today’s intelligent information management solutions offer ways to automate the time-consuming and often boring document-driven processes within a business.
One of the key drivers in this automation is the use of artificial intelligence and machine learning. AI and machine learning have the ability to reason and discover meaning as well as learn from past experience. Moreover, artificial intelligence systems can easily churn through lots of information to recognize patterns and categories in the data. That ability is put to work to enable new ways to search, find, use and manage information, and add automated workflows to document management processes.
But, artificial intelligence and machine learning mean very different things to different organizations and people. To us, they mean something that really drives the development of the industry but is at the same time very simple and easy for the end users to use and benefit from. Our objective is to allow customers to intelligently find and manage the critical data they need to make smart decisions.
Jayson deVries, our artificial intelligence expert in the Product Management team, lists three major, emerging themes related to AI and machine learning:
Artificial intelligence should not be complicated for the user
We believe in providing intelligence services out of the box, so that you or anyone in your organization do not need to be a data scientist to use it in your business.
The thing that differentiates M-Files AI from others is the usage of machine learning behind the scenes. The users do not need to change how they do things or take extra steps to train the system. Just by observing what the users are doing, M-Files AI will then begin to assist them automatically and will continue to improve the more it is used. We have worked hard to provide a tool with this self-learning capability, so that even an SME can benefit from AI without needing to hire expensive data analysis expertise.
Some products offer fantastic, powerful AI capabilities, but they require a deep understanding on how to carefully train, test, tune …and repeat. If you attempt to implement this without the correct skills, it can turn into an expensive endeavor that doesn’t end well, with the product taking the blame. Then there are products that offer out-of-the-box by delivering pre-trained “canned” AI. These do have merit but must only be used in their intended contexts – a fact that, if overlooked or misunderstood, will cause disappointing results.
AI needs to be tested with your data
There has been a recent onslaught of “intelligent” products and services on the market, however some of these are not really that intelligent at all. It is important for customers to be able to see what’s behind the scenes and ideally understand how well an AI solution really works with their own data, not just demo data.
To be able to execute testing with real data before the decision to buy, the AI solution needs to be easy enough to implement. If it takes too much effort to get the artificial intelligence to work, it can be totally impossible to try it out with real data.
On the surface, an AI feature can sometimes appear rather mundane or obvious to a user. All the clever algorithms and mechanisms are invisible to them and they just see something happen automatically.
Jayson explains: “I’ve recently been using the classic duck analogy: ducks are swimming around calmly, but a lot of impressive work is happening below the surface to make sure the duck gets to where it needs to go. Extending that analogy further, some vendors are now selling ‘rubber duckies’. They look and quack like a duck – if you squeeze them just right – but they’re actually hollow with nothing below the surface.”
Cloud leads the way, but on-premises AI is an opportunity
Cloud services and SaaS have been around for some time now, but certain customers and industries have not yet embraced them fully. As long as AI services are only available in the cloud, there can be significant dollars left on the table. So, even though it might look like a bit of an old-fashioned way, there is a real opportunity for premium, fully on-premises AI backend solutions.
A related, emerging technology is so called Edge AI where AI is executed directly on the device so that the data never even leaves.
We at M-Files have taken a cloud-first approach but are also working toward offering many AI tools for on-premises installations.
M-Files intelligence services
We are launching a new tool, M-Files Smart Classifier, as an add-on feature in November. M-Files Smart Classifier is based on the approach described above, where the product learns while it is used, and it provides suggestions for what type of a document a certain document is.
In addition to the new M-Files Smart Classifier, we offer other intelligence services, for example M-Files Repository Sensor, which helps crawl through large amounts of data and identify business critical information, such as personally identifiable information, from clutter.
These services are all part of offering increased automation and improved efficiency in office work.
Our aim is to develop the services so that we have three steps towards intuitive, intelligent information management:
Step 1: Provide meaningful metadata suggestions
- Use AI to analyze and understand the content users are managing.
Step 2: Separate business critical data from dark data
- Crawl through data repositories.
- Find relevant data, make it visible, and attach it to business processes.
- Remove information clutter.
Step 3: Offer users data that is relevant for them now
- Understand who the user is.
- Provide proactive information that is relevant today (not yesterday or tomorrow).
- More relevant search for the user.