Blog

Exploring Enterprise Search and Knowledge Graphs

The ability to search across various systems is a growing need with many of our clients. Content is created by different teams, in different systems within an organization and a broad array of consumers of said content require the ability to find what they need. When faced with a search project on this scale, two highly adopted solutions are enterprise search and knowledge graphs. In this blog, I’ll discuss the benefits of enterprise search and knowledge graphs and how they work together.

Enterprise Search

What is Enterprise Search?

Enterprise search pulls content from multiple disparate systems into a search engine using connectors that extract the text and metadata from various systems, and indexes the content for use in search interfaces. A search index is an optimized structure of the content to improve the efficiency of search. The initial population of the index takes time but a carefully curated search index creates a richer search experience by surfacing more relevant results and enabling the use of keywords and synonyms.

Why Enterprise Search?

Desktop with example search results pulled using a knowledge graph.An enterprise search solution is a one-stop-shop for your organization’s search. Search engines often provide access control “out-of-the-box” as well as pre-built connectors for ingesting content. These speed up the initial deployment process and help ensure document security policies are met. Likewise, an enterprise search system increases productivity by allowing employees to find and share information that would otherwise need to be reinvented. The addition of action-oriented results and search design best practices are simple steps that also greatly improve user experience and enable users to act on information quickly. Overall, an enterprise search solution reduces employee time spent looking for information in different systems, improving productivity and saving resources. 

Knowledge Graphs

What are Knowledge Graphs?

Knowledge Graphs combine content from different systems, creating a network of entities for your organization. 

A network of entities? What does this have to do with search?

Enterprise knowledge graphs are a growing trend in the industry where all entities are connected and stored in a relational web. The entities are your organization’s departments, employees, clients, projects, deliverables, or other domain-specific items. Knowledge graphs are built by integrating with other systems, pulling in the information needed to describe each type of entity. This knowledge network enables organizations to not only search these entities but also explore how entities are related.

Why Knowledge Graphs?

The primary benefit of a knowledge graph is that entities can be connected regardless of the type or source. Whereas search is like a funnel, allowing you to search and filter content from different sources, a knowledge graph connects entities from different sources and allows business users to ask complex questions across the connected systems. This connectedness also facilitates knowledge discovery as users are presented with relevant content that they may not have known existed. Furthermore, knowledge graphs enable entity recognition for search knowledge panels and are the next step in your taxonomy development as your organization grows and describes its domain. As your organization adjusts to meet user needs, a knowledge graph has the flexibility to support more use cases than just search.

Enterprise Search and Knowledge Graphs

An ideal search solution is just like a holiday dinner, a combination of Ma’s sweet potatoes and a perfectly cooked turkey.

A combination of enterprise search and knowledge graphs gives users a cohesive search interface to find and explore their organization. Use enterprise search to curate content from different sources and deliver an accurate, efficient search experience. Extend enterprise search with a knowledge graph to give more context to search results, provide support for natural language questions, and enable discovery through relationships.

Alternatively, start with a knowledge graph to explore your content–focus on interpreting relationships and recommendations. Add enterprise search once you need to provide full-text search and faceting to users.

Do you want to explore your options or need help implementing the next phase of your search project? Reach out to us and we’ll be happy to help build the best search for your organization. 

 

James Midkiff James Midkiff James Midkiff is a developer in a range of languages and associated technologies. He is focused on system design, implementation, and integration with both open source and COTS tools, as well as the connection between these tools and their business users. More from James Midkiff »