How a Business Knowledge Graph Works

Original article was published on Artificial Intelligence on Medium

How a Business Knowledge Graph Works

It’s the engine that makes an AI advisor feel like a data expert in your pocket — find out what it does.

Using an AI advisor is simple — you plug in your business data and ask for insights. But there’s something more complex going on under the hood.

An AI advisor’s goal is to deliver the best response to every question. Responses personalized to the user and their business, and enriched with multiple data sources to give the full picture. This takes something more than just presenting insights closely matching the user’s search terms.

It takes a Business Knowledge Graph.

Way to connect information

A quick way to see how a Knowledge Graph might work is to look at Google. Say you search something as simple as ‘iGenius’. You don’t just see a link to our website, you see what they call a Knowledge Graph Panel on the right with different bits of information — like our office location, reviews, company description or more.

This is the result of a Knowledge Graph — a system that connects pieces of information and helps understand the relationships between them.

A Business Knowledge Graph makes meaningful connections between your data sources

We take this concept and apply it to businesses. Our AI Advisor’s Business Knowledge Graph organizes a company’s raw data and, with Machine Learning algorithms, makes meaningful links between multiple data sources based on your context. This helps make every response thorough, relevant and personalized.

How it works

Let’s look at how a sales department in a fashion retailer might have one set up. There are two steps:

Step 1: Lay the foundation

You start by laying the foundation — the base on which a Business Knowledge Graph grows and shapes itself over time. This covers three elements:

  • Your Business Framework. What databases do you need to connect? What insights do you need to extract from your existing data ecosystem? In this case, it could be CRM Data, Customer Care Data, ERP Data or General Market Data. The information needed to get started.
  • Your User Profile Definition. Who’s going to use the Advisor? What are their roles and duties? What do they need to see? For a Fashion Retailer, this could be the Head of Sales, the Area Manager and a Sales Agent.
  • Your Business Specifics. What’s your industry or business area? What jargon and keywords go with it that the AI advisor would need to understand? Here, it could be ‘Clothing Ranges’, ‘SKUs’, ‘Promo Deals’, ‘Loyalty Schemes’ etc.

Step 2: Grow it by using it

Once the foundation’s set, the Business Knowledge Graph is ready. If data is the information, the Business Knowledge Graph is the brain that absorbs, processes and makes sense of it — learning and getting more sophisticated over time.

Visual demo of how a Business Knowledge Graph works

So, as you ask questions like ‘Which stores are performing the best?’, choose topics the advisor suggests to explore and interact with push notifications, the Business Knowledge Graph grows.

It learns what you like to see, what you don’t like to see, and pinpoints the most relevant insights — making everything more tailored to you and your business.

See an AI advisor in action

The Business Knowledge Graph is just one element that makes crystal, our AI advisor for data intelligence, tick.

Putting all your KPIs in one conversational interface, crystal guides you through your data, notifies you about important changes and lets any team member make data-driven decisions from their phone.

Click the button below to try a free crystal demo.