Predicting what people need when they need it: That’s the key to better care adherence, increased financial stability, and simply meeting life’s most important goals.
This has historically been a challenging task, given the limitations of technology in the benefits space. But with the adoption of artificial intelligence, machine learning, predictive analytics, and intensive data analysis, much more is possible. The Data Science team at Evive was meticulous in choosing exactly the right approach to solve these problems in benefits engagement—problems like not choosing a relevant health plan, or poor utilization of benefits that have the ability to change a person’s life.
We assessed the exact challenges and identified what was needed to overcome them.
The knowledge graph
Evive leverages next-generation analytics to help employers provide personalized benefits messaging to employees. Through our machine-learning and deep-learning algorithms, we can anticipate and identify what people are likely to need and when—for example, if they’re at risk for certain medical conditions (think back surgery, high cholesterol, hypertension, cardiovascular disease, and so on) and which resources they’ll benefit from.
With such a vast amount of medical data and models on hand, how do we tie them all together? This is where the knowledge graph comes in, as every piece of information from all of our algorithms is mapped together. Each data point is connected to other relevant ones in a dense network so they can quickly be retrieved in a format that’s easy to understand.
This is essentially how our brains work—via encoded memories, context, and world-models.
A dense network of context connects ideas
Think back to the bedroom you had as a child. Chances are you get a quick mental snapshot of your room. Your brain probably recalls an intricate amount of detail—almost as if you were transported back in time and looking in. In an instant, you may remember the room layout, your bed, a table next to your bed, and much more. Maybe even the feel of the carpet against your feet.
How did you remember all that in a flash of a second?
Your brain stores and retrieves memories by tying them together with a common thread—a context. As new memories form, they tie into the already existing elaborate network of information in our brains. Memories that are recalled often, and those with robust connections through multiple points of context, end up entrenched in our minds. New material with weak ties to an already existing memory network ends up more easily forgotten.
We also have models of the world in our minds—heuristic and empirical. We may not necessarily know Newton’s Laws of Motion, but we sure do know what goes up must come down! And we act based upon those memories, contexts, and world-models.
How AI and the knowledge graph mimic our minds
The field of artificial intelligence has long tried to recreate how the human brain works. In 2012, Google released the first knowledge graph, a database of information available on the Internet. Every data point that exists online connects to the searches we conduct. This database was built to simulate the process of creating and recalling memories in our minds.
Now, data scientists at Evive are using similar technology to enhance experiences. Any interaction with the user is not only personalized, but context-aware. Let’s say someone searches for a back-pain procedure—this is evidence that the person is interested in information regarding back pain. From the knowledge graph, we can gather information on the various causes of back pain and the complications that can arise from it. Moreover, based on the person’s historical medical data, we can derive other relevant information that should be provided.
In a nutshell, our AI models + knowledge graph will:
- Identify predictors, like osteoporosis, arthritis, knee pain, etc.
- Comb through numerous employee profiles on the database
- Select and analyze information that may indicate risk
- Quickly return accurate results in a user-friendly format
Moving benefits forward
These are but a few examples of how our knowledge graph is advancing the field of benefits. The opportunities are limitless, and we’re on a mission to solve those key challenges in benefits engagement.
As we keep tapping into this innovative database, we will continue to stay ahead of the curve: connecting people with the resources they need most and meeting the evolving needs of our customers.