A trend for information self-service raises the relevance of complete applications vs. tools
In our discussions with customers, we have noticed a trend that is significantly impacting the way that semantic solutions fit into the corporate landscape. In the “good old days,” an employee with a question would submit it to an in-house team of information experts who would use a set of expert tools to respond (by way of feedback report or list) to the initial enquirer.
While this is still common in many places, we are noticing a tendency to rely more and more on self-service solutions that put the employee in the position the end user, accessing solutions that attempt to address the bulk of incoming questions with web-based solutions. Welcome to the world of the intelligent search engine.
There are a number of factors at play here. More and more users feel confident performing searches on their own. Encouraged by the (often misleading) impression that, because they know how to use their favorite search engine, they think that the same type of search is also effective on a professional level. Rather than consulting an in-house team of experts, the average user would just go ahead and look for themselves.
At the same time, solutions that provide quick access to relevant information–i.e. Intelligent search engines– are becoming more and more powerful and intuitive. Finally, budgets for information departments are tight and frequently reassessed.
The result is that more and more users who are experts in their specific domain – such as biology, law or engineering – find themselves exposed directly to search and analysis interfaces without the reassuring presence of an information professional to help guide or improve the experience. This places a major responsibility for performance on the part of the technology solutions and tools. While it may be acceptable for private users to engage in search without a more detailed understanding of the principles behind a search engine[1], the same is not true in an enterprise environment. In a corporate setting, where incomplete or misguided search results may have a much larger impact, the pressure to have appropriate tools is, of course, much higher.
As one project sponsor put it: “The aim of the solution we want is not to make 5% of our staff 100% smarter, but to make 80% of our staff 10% smarter.” After all, many users hardly ever use even Boolean search. As a result, they would not be able to gauge the full potential of thesaurus-based indexing or more advanced technologies that are available for improving search efficiency.
Read also the article about Natural Language Processing Systems in Artificial Intelligence.
Intelligent search engines at Expert System
At Expert System, we translate this into an increased effort to develop not only the underlying information processing engine, but also the complete application. We know that we have to be able to address the fact that intuitive interfaces and effective document processing tend to hide the complexities of what goes on “under the hood” when it comes to an intelligent search engine.
Our Biopharma Navigator is a perfect example of this approach. The information that professionals in the life sciences and pharmaceutical sector need access to is too complex, heterogenous and distributed to be digested quickly without the help of both an information professional and intelligent search technologies.
Biopharma Navigator is a web-based solution that allows domain experts who are not information professionals to conduct a set of typical scenarios such as searching for experts, news about a therapeutic area or competitor activities. Results, enriched by typical synonyms behind the scene, are presented in rich dashboards that provide immediate access to the information they’re looking for.
Figure 1 – The Biopharma Navigator provides users with quick and intuitive access to more than 40 million documents.
Today, our Biopharma Navigator is used professionally by more than 1,000 industry experts to quickly and easily access the information they need.
In a similar vein, our Analysts’ Workspace software implements the same idea: Putting powerful semantic technologies in the hands of people who need quick and comprehensive results (without having to start from scratch for each new search).
Not everyone can be an expert in text analysis: Some of us need to be able to cure cancer, make good investment decisions or assess political risks and opportunities while leaving the analysis of the required sources to the experts. At Expert System, this is our mission.
Stefan Geißler
Senior Cognitive Scientist, Expert System
[1] Vgl. http://www.blogs.uni-mainz.de/ifp/files/2013/08/Suchmaschinen_Management_Summary.pdf
