Day 4 - Turning uncertainty into clarity

Yesterday I was mentally stuck trying to flesh out this knowledge graph recommender system. I created flashcards for each of the two goals that I was testing: 1. Build a Consistent Fitness Routine and 2. Stay Mentally & Socially Engaged.

But I kept at it and I think I reached a point where I can build a system, test it out, and then go back to the original design to improve upon.

Here are some steps I need to take for building out the system:
-Define if flashcards require triggers
-Define which concepts the card teaches and if the card has constraints
-Define which goal, phase, week, and day the card belongs to
-Turn my features into cards and tag with above as well (e.g., journaling, food logging, workout generator, voice agent, meal prep event scheduling)
-Create a nodes and edges glossary based on the above and feed all cards to a knowledge graph
-I think I have to create another knowledge graph just for tracking user data but I am still not sure

In the short term, I see the knowledge graph capturing information from the user and in the longer term helping us prioritize what we show users based on real engagement. But yeah I am still a bit unsure how the knowledge graph comes into play beyond capturing relationships. I have not come to a confident state about the final recommender system since the above steps I defined could be implemented without a knowledge graph.

On we go...