Can a combination of SharePoint, Bing and AI be the next wave in knowledge management?
Microsoft Ignite 2019 in review: Azure Arc, HoloLens 2, Edge, Quantum and Teams
With around 26,000 attendees at Microsoft Ignite 2019 in Orlando, the company has announced dozens of new features, products, and updates.
More about artificial intelligence
The transition from Microsoft from Office 365 to Microsoft 365 is almost complete, with the operating system, management tools and productivity platforms brought together in one subscription, supported by the connection layer that is the Microsoft Graph. At its recent Ignite event in Orlando, Microsoft announced what it describes as the fourth pillar of Microsoft 365 – Project Cortex.
Perhaps best seen as the latest version of knowledge management, Project Cortex is a way to bring together key aspects of Microsoft 365 to help you find answers to questions about your organization and the work it does. Building on well-known tooling, it is perhaps best to regard it as an evolution of the Microsoft Search tools available through Bing that add work queries to your browser.
Microsoft Search focused on context-based searching, using the tool you are searching to refine the underlying search. A search in Outlook would be mainly for e-mails and in Word for documents. However, it is still an intranet search tool, although it is better at providing the answers you are probably looking for.
Go beyond Microsoft Search
Building on that basis, Project Cortex continues, using machine learning to generate content based on documents and corporate social network conversations, constructing what might be best described as a “business Wikipedia.” By using elements from Azure’s Cognitive Services and the Microsoft Graph, content can be linked to key figures (such as in Microsoft Search) and delivered in channel-appropriate formats.
Most interactions with Project Cortex will go through what Microsoft calls “Subject Cards.” These appear in suitable places – initially in Word, Outlook, Sharepoint and Teams. From the first view we’ve had, as well as showing important information about content and people, they let you follow specific topics and suggest any edits. Topics are reflected in the ‘People cards’ of Outlook.
For example, you can search for people related to specific projects in Outlook, with acronyms and definitions in Word, and quick responses through bots and “Adaptive Maps” in Teams. Project Cortex applies existing role-based access controls to content, so that the content of a commercially sensitive project is only visible to project team members. Other content can be limited to the summary form only, so that detailed content is preserved for only those with the correct access rights.
Project Cortex uses AI to automatically organize content, deliver topic cards, topic pages and knowledge centers in Microsoft Office, Outlook and Teams.
Image: Microsoft
Add machine learning to knowledge management
Project Cortex is not just for Office documents. Using Azure’s Cognitive Services, it can use image and text recognition to work with scanned content, images, and other file formats such as PDF. It can even use rules to define form structures, so that key information can be extracted from scanned forms and other common document types, so you can build a model of where projects spend money by parsing purchase orders and invoices. Extracted information is used as metadata to provide context around documents and to help users find the content they need.
You are not limited to structured document types. Another Azure Cognitive Service, LUIS, forms the basis of the Machine Teaching of Project Cortex. Here you can build new document models that search for key terms, which can, for example, classify contracts that differ from contract to contract, with different content and different formatting. Once a model has been trained, it can be used throughout your document archive, improving search and increasing the underlying knowledge model of your organization.
SEE: Microsoft HoloLens 2: a guide for insiders (TechRepublic download)
Project Cortex comes with a set of connectors for third-party repositories, so you can take advantage of investments in enterprise content management tooling and in older knowledge management platforms. The first set includes support for Azure Data Lakes, ServiceNow and Salesforce. With a recording API, other suppliers can quickly add their own connectors.
There is an element of content control in the tooling: if, for example, the generated content for a business abbreviation or project name is thin or inaccurate, internal editors can fill this in with more information. It is not necessary to learn new skills here, because the underlying content platform is based on SharePoint.
Expand the Microsoft Graph with Project Cortex
It is important to understand that none of the information generated by Project Cortex leaves the Microsoft Graph. Instead, it adds new metadata to your documents, using an automatically generated taxonomy, the Managed Metadata Service. Because it is a managed service, you can add your own metadata or create new tag definitions. The resulting metadata, stored as knowledge entities in the graph, is used to define the topics that yield Project Cortex results.
An important driving force behind Project Cortex is demography: the bubble of baby boomers leaves staff quickly when they retire and takes knowledge and skills with them. Much of that knowledge is not completely lost, but it is trapped in the many terabytes of files with documents that have built up in our business systems. We need a way to transfer that knowledge to new generations of employees in all our companies. By using Project Cortex to capture and index that information, and with AI tools to create ways to navigate and extract that data, we can turn business document storage from a memory hole into a dynamic learning environment.
SEE: Management of AI and ML in the company 2019: technical leaders expect more difficulties than previous IT projects (TechRepublic Premium)
Once that information is available to everyone, it can be used as a basis for dynamically generated frequently asked questions and for business search results, with just-in-time access to essential content that not only helps in transferring skills, but can also be used to scales-expanding companies. Project Cortex not only collects knowledge from documents; it can also work with tools such as Yammer and Teams to extract the implicit knowledge in our companies – the things that everyone knows informally and that are passed on by the person you are asking instead of reading a business handbook.
This is where much of Machine Cortex’s machine learning comes in handy, which helps in extracting and categorizing that informal exchange of social corporate networks, as well as more formal documents. It is an ambitious step for Microsoft, because previous efforts to implement knowledge management tools on a large scale have not gone well. Project Cortex’s mix of Microsoft Graph and machine learning may finally bridge the gap between data, information and knowledge.
Weekly newsletter from Microsoft
Be Microsoft’s insider for your business with the help of these Windows and Office tutorials and the analyzes of our experts on Microsoft business products.
Delivered on Monday and Wednesday
Register today
