Show HN: We created a tool to visualize scientific knowledge

keywords.groundedai.company

10 points·ml_more··5 comments

I posted our project a few days ago in a thread about Vitamin D. We’ve made some UI improvements people might find interesting.

Open articleView on HN

The implied relation between nodes seems to be "is related to" or "is associated with" - the same kind of babble you see in PubMed research where authors don't want to go out on limbs and make causal claims.

I would argue that a rich knowledge tool should spend a lot of time teasing out the nature of those relationships.

Having said that, this class of visual tool is going to see a lively future.

I agree, it's currently very superficial. It would be great to recognise and show more about the specific nature of the relationships.

What kind of thing would you hope to see here? A textual summary of the relationship? Or perhaps there is more that can be done with shapes and colours in this area?

This is a rich topic and great question, one to which we should all try offer answers.

From my experience, I work with topic maps - extremely rich knowledge graphs, and an ISO 13250 standard, and I work with NLP, which entails a lot more detail than just talking about relations: they are first class citizens (subjects) in a graph, not just labeled arcs. I spoke about them here [1]

[1] https://www.slideshare.net/jackpark/lbd-tm2

What are the colors of the nodes representing?

One of the creators here.

The keywords are grouped such that keywords that occur together often are in the same group. The colour represents this grouping.

There may be more useful groupings we could give or allow the user to choose between. We would be interested to hear any ideas for that

Reply on news.ycombinator.com