Press Releases | Starmind

Starmind Awarded Patent for Cutting-Edge Artificial Intelligence

Written by Stijn Vermeeren | Apr 2, 2019 10:00:00 PM

The value of most knowledge management solutions is that they make existing documented knowledge more accessible. Accounting for about 20% of all employee knowledge within an organization, these tools are pretty useful. 

Starmind goes several steps further. Our platform gives organizations access to the other 80% - the implicit knowledge of its employees which can’t be found in any document. 

It’s this cutting-edge technology which has been recognized in our recent patent. 

What is the patent for?

Our platform is formally described as a “machine learning approach for query resolution via a dynamic determination and allocation of expert resources”

In layman’s terms, we help customers access the collective knowledge of all their staff across the world.  

In this piece, we’re going to examine several key aspects of Starmind, as recognized by our patent. These are: 

User queries get routed to the correct expert, who can then provide a solution.
The determination of who is a suitable expert is done using intelligent algorithms.
Starmind's algorithms learn in a continuous and dynamic manner who is an expert on what.
Starmind understands if a similar query was asked and answered in the past.
Starmind can distinguish between who is interested in a topic and who is actually an expert on a topic.
Starmind can learn and adapt to organization specific vocabulary.

Queries are routed to the most relevant expert

Employees are regularly faced with tricky problems or questions they can’t immediately answer. However, in a large enough organization, it’s almost certain someone has the solution. 

A crucial aspect to Starmind is its ability to connect employees with the most relevant and qualified company experts on a given topic - no matter where in the world they’re based. 

Once Starmind identifies the most useful expert(s) for a specific question, they are contacted directly for a response. On average, Starmind solves over 95% of all questions posed - 70% of them from a separate location! 

Intelligent algorithms decide who the experts are

It’s hard for individuals to accurately rank their level of expertise in different areas against colleagues they’ve never met, seen, or heard of. Additionally, just because Maria says she’s an expert in Python programming doesn’t mean she would be considered an expert by others. 

Starmind uses advanced, intelligent algorithms to alleviate this problem. The platform weighs up  a host of different factors and objectively identifies the experts who are best qualified to answer each individual query. 

Dynamic and continuous adjustments to “expert” status 

But what constitutes an “expert” changes over time. For example, if two people demonstrate equal knowledge on a subject, but one is much more responsive to questions, they might be ranked higher of the two. 

Starmind evaluations are constantly optimized in order to present the most valuable and qualified minds right now - this is something no other technology can deliver. 

The result is that queries are always answered by the most qualified experts within the company. This means faster and more effective query resolution. The more questions that are asked and answered, the more performant the system and its users become. 

Recognizing repeated questions and answers

It’s easy to ask the same question in a hundred different ways. A user might even think two questions are different where in fact the problem (and resulting solution) are the same or near-identical. 

Starmind uses advanced algorithms to understand when same/similar queries have been asked in the past. The platform will automatically suggest those similar questions (and their answers) to the searcher, helping reduce the amount of time spent problem solving. 

This is also efficient for the experts who are not required to repeatedly provide the same answer over and over again. This is just one of many efficiencies that Starmind grants its users. 

Distinguishing experts from interested contributors

Any employee intrigued by a query or topic might offer their two cents - and with all the best intentions, theirs may not be the most valuable answer. Starmind is capable of determining experts from interested parties and indicating this to the asker. 

This allows searchers to give more attention - if they desire - to the expert answer, and ensure the problem is resolved fully. 

Adapting to organization-specific vocabulary

Every company in every industry uses “niche” terminology: words and phrases which are not otherwise used in daily speech. In some cases, there is even a way of speaking or asking questions which is different from normal. 

Starmind picks up these trends, becomes comfortable with this specific vocabulary, and learns to address it seamlessly during operation.  

Conclusion

There’s a lot that goes into making Starmind a highly performant and valuable platform. Our recent patent recognizes the innovation and bleeding-edge technology we have created, as well as the unique amount of insight and value it offers to clients. 

To learn more about Starmind and how our technology helps businesses utilize their collective knowledge, visit our How it Works page.