Knowledge Portals & Why They Matter in Business

As enterprises amass more data and content, the need for a system that can keep up with all this content becomes more urgent. 

Volume is not the only thing that poses a problem; it is also the variance in file types, the disparate systems that files are saved under, changing authorship within a company, and so on and so forth.  

A Knowledge Portal offers the solution to these headaches via a rich ecosystem of semantic technologies, data management tools, and user applications.

Our partner, Enterprise Knowledge, defines a portal’s infrastructure typically being “a content management system, enterprise search, and web front end and may also include a knowledge graph, metadata hub, and taxonomy management system.”

The knowledge portal is a network of the tools an enterprise likely already uses PLUS the advantages of semantic technologies. At Semantic Web Company, we provide the tools to obtain the semantic advantages with advanced text mining, taxonomies, concept tagging, knowledge graphs, semantic search, and recommender systems. 

One one end, data is stored in its respective repositories and gets treated by semantics to create active metadata. The active metadata is then contained in a graph database to link all the pieces together with their context so they can be better retrieved by the search engine on the other end.

Altogether, knowledge portals connect various layers of an enterprise to ensure that data and content is being processed and used to its fullest potential. 

As far as content production goes, much of our time is spent extracting content building blocks from different silos, and then assembling them in a way that makes the most sense in a specific work context. All of this is largely done manually, as enterprise content infrastructures are not yet optimized to meet these requirements.

Say you want to create a product sheet for a laptop that your company has just launched. You need strong marketing copy for the headline and introduction, specs and features from Product Development, graphics and logos from Design, prices from Sales, contacts for Presales, and the list goes on. There are many aspects that go into a simple 1-pager with many points of authorship, that the project leader often has to chase after because all this information is contained in silos or under different naming conventions. 

A knowledge portal can provide a few unique features to ensure these processes become more automated and efficient:

• • Metadata that is machine-readable and meaningful outside of departmental silos
  • Controlled vocabularies to understand and process content across roles 
  • Knowledge models that describe the semantics of the metadata and can classify it automatically
  • Digital assistants that help to link content across silos that belong to each other on the level of meaning, not just at the level of terms used (aka “recommender systems”)

Being able to find things more quickly is a chief advantage of a Knowledge Portal, and our partners at Squirro say that cognitive search employed by Semantic AI, reduces search time by up to 90%. “Cognitive Search delivers results that are more relevant to the user, boosting productivity and reducing costs…there is finally a replacement for the obsolete keyword-based enterprise search that depended on users knowing exactly what they were looking for, as well as the right keywords to find it.”

In the case of our product sheet, the features of the Knowledge Portal that are powered by advanced search capabilities, would help the project leader find information quickly so that the process of creating and publishing it is more seamless.

At Semantic Web Company, we have incorporated a Knowledge Portal into our own knowledge management strategy. The team is revelling in “drinking our own champagne,” so to speak – where we can all access the information that has been collected over the 15+ years of being a company with ease. 

The Knowledge Portal first started out with the following indexed pages:

• All chapters of the Knowledge Graph CookBook (149, GDrive)
• The PoolParty Help Pages (957, Paligo)
• The SWC collection of analyst reports (222, GDrive)
• A selection of SWC Marketing collaterals (34, GDrive)
• All pages of the poolparty.biz website (128, WordPress)
• Customer use case descriptions (48, Confluence)
• Presentations from KM World 2021 conference (45, GDrive)

When the portal was first being built, our knowledge engineers tagged all these documents and created a taxonomy that supports the whole architecture of the application. Today, the portal still continues to be updated with new articles and information, indexing them on a daily basis. 

Employees are using the portal to offer customers better services by matching their profiles to other business scenarios, speed up time-to-market by referencing analyst reports with product pages to create product launches, and so on. There are numerous points of value that the portal provides with a key benefit being that all this information is in one place, allowing users to make connections they would not have spotted otherwise. 

In a “post-pandemic” era, where hybrid formats are the new normal, enabling employees with the information they need in a centralized place is especially critical. Semantic Web Company has successfully implemented this internally, and is working with customers to help them build their own Knowledge Portal ecosystem. 

Microsoft Docs and HealthDirect Australia are two examples where this is already working, with many to come in the future.

All these capabilities point towards better search and findability of business objects. However, an important distinction can be made between search and recommendation – and it is the latter that separates a Semantic Knowledge Portal from other solutions on the market. This is not to say that users can’t benefit from semantic search in their portal, but to add that the semantic search capabilities are enhanced even further with the additional recommender system.

A semantic recommender system can create “pairings” based on user activity or similarity of content, such as a ‘Show similar documents’ functionality while searching for relevant documents.

We have all had the experience whereupon purchasing an item on an ecommerce site, we are immediately recommended the same item but maybe in a different color. Most people don’t need two of the same couches in their living room, but they could benefit from being recommended a coffee table or a lamp. A semantic recommender system does exactly this: it can understand what you are looking for and even suggest things that can help you further.

Taken in the context of a Knowledge Portal for an enterprise, an employee searching for a value proposition summary for a prospect could also benefit from being recommended an article that lists all the times that the company was mentioned in Gartner publications, a price list, use case material with the relevant contact who worked in building a previous PoC, etc.

The original search query used the keywords “value proposition,” but through identifying the context of these documents, the recommender system could suggest related documents that the employee could then offer in a comprehensive package to the prospect. 

The Knowledge Portal – powered by the recommender system – increases the findability of what the user is searching for and helps them take advantage of new information that is just as useful to the original search intent. 

With these approaches, an enterprise can feel better prepared to offer more refined services to their customers.