Growth Series BLOG

Its a Premium BLOG template and it contains Instagram Feed, Twitter Feed, Subscription Form, Blog Search, Image CTA, Topic filter and Recent Post.

All Posts

What is Semantic Search?

Semantics is the study of meaning. So a semantic search is one which “seeks to improve search accuracy by understanding searcher intent and the contextual meaning of terms … to generate more relevant results.”  Human knowledge and the language that humans use to communicate that knowledge are ambiguous and messy. 

A semantic search is one which “seeks to improve search accuracy by understanding searcher intent and the contextual meaning of terms … to generate more relevant results.

Take the term” diamond,” for example. It could refer to jewelry or sports. If I am searching for “mercury,” is that the planet, the element, or the car? We see the problem in business every day. People use the same term to describe different things, and they also use different terms to describe the same thing. They use vague terms such as “searching for a solution” or “looking for an agreement.” 

The source of search problems - not the technology

Most organizations have problems with search not because the search engine is inherently flawed. Keyword search engines do what they are told: they look for word occurrences. Humans and human language is the bigger problem. For those organizations mature and disciplined enough to utilize the tools and technologies currently available (such as metadata management in SharePoint) it is possible to create very effective search experiences.

However, most organizations do not have their acts together when it comes to enterprise taxonomy and information architecture, which are the core foundation for these processes. In fact, a recent survey of practitioners and executives from industry (283 responses from enterprises of various sizes) found that over two thirds of respondents (68%) rated their organizations as unmanaged or limited in terms of their maturity in enterprise taxonomy. (The choices were “Unmanaged,” “Limited,” “Pervasive,” “Proactive/Optimized,” and “Strategic Business Asset.”)

Without the core of intelligent classification, it is not possible to be successful in content organization, and therefore search will not be as effective as it might otherwise be. However, the new “semantic technologies” can go a long way toward compensating for these structural deficiencies.

Where technology can help

Semantic technologies can interpret the aspects of search  that humans are good at and computers are usually bad at: meaning, nuance, and context. Computers are good at working with long lists of information, large numbers, and calculations -- things that can be dealt with through pure logic. Computers have traditionally had a very difficult time with language and interpretation that relies on subtle differences in context or usage. But, the times they are a changin’.

If metadata is lacking because a coherent taxonomy has never been developed, pre-processing the data can help achieve clarity. For example, if a user is looking for content that pertains to the “Acme” company’s “Deluxe Widget.” Many documents contain “deluxe” and many that contain “widget,” and even more that contain “Acme.” But if the three terms are together, then you can be assured that this is the right product information. 

Semantic technologies have the ability to subject the queries to all sorts of pre-processing manipulations which can effectively identify the user, understand the term, interpret that term based on a use case, and associate terms that are automatically discovered.  We can use query rules to change the nature of a term to align more closely with what the user is asking for. This effectively means that the search engine can recognize the problem that the user is trying to solve and can present results that depend on the identity of the user and the nature of the tasks. 

Semantic search-based applications represent the next stage of search, tuning the results to specific end users and getting people the information they need in the context of the goals they are trying to achieve. 

Learn the crucial roles that taxonomy and metadata play effective enterprise content and knowledge management systems in Searching for Gold: Harnessing the Power of Taxonomy and Metadata to Improve Search.

Seth Earley
Seth Earley
Seth Earley is the Founder & CEO of Earley Information Science and the author of the award winning book The AI-Powered Enterprise: Harness the Power of Ontologies to Make Your Business Smarter, Faster, and More Profitable. An expert with 20+ years experience in Knowledge Strategy, Data and Information Architecture, Search-based Applications and Information Findability solutions. He has worked with a diverse roster of Fortune 1000 companies helping them to achieve higher levels of operating performance.

Recent Posts

How to Design Omni-Channel IVAs That Humans Love To Use

During our webinar, "Omni-channel Virtual Assistants - The Next Frontier in Voice and Text for Customer Service guest speaker Chris Featherstone, who leads business development for AI and speech recognition services at Amazon Web Services (AWS) and Seth Earley, CEO of Earley Information Systems (EIS) discussed ways to design omni-channel virtual assistants to optimize their use across voice and text. When supported by an appropriate information architecture and designed with a deep understanding of the customer, virtual assistants can access enterprise knowledge efficiently, saving time and money. The key to success is to structure the underlying information so it can be retrieved and used by any channel, including humans, to deliver the responses that customers need. Here's a recap of the conversation.