Comment by purephase | Hacker News Reader

Comment by purephase | Hacker News Reader

Back in my sleep-deprived days after my oldest was born I had an app idea similar to this, but with an added dash of AI/machine learning.

Imagine a bunch of parents using a free/relatively inexpensive application, tracking feeding schedules, sleep, food intake, play etc. and then ranking each day as a thumbs up/down.

Taking that data, properly anonymizing it and then comparing it across babies with similar characteristics (location, age, siblings etc.) you could suggest a daily schedule (with alerts) that would trend closer to a thumbs up.

Started with a prototype, but it never amounted to much. I just imagine the collective value of some of the data managed by these applications and the potential to ease the extremely challenging (but rewarding!) experience of being a parent.

That's generally what we aim for. I'm co-founder of Baby Manager, it's free a baby tracker app. You can track all things you mentioned. Our plan is to make use of this data in a way that parents can understand their baby better. They can already see patterns on Timeline view or charts but definitely there plenty of ways to get much better insights out of it.

I thought about this along the way. It would be really cool to have some alert that says, "Baby is probably ready for a nap" or "Baby probably needs a diaper change", but the ML side of that is far outside my domain. Did you get far enough to judge whether or not such a thing would be very feasible, given how babies can develop so drastically differently?

I didn't get far enough, unfortunately. This was a number of years ago, and not the panacea for ML/AI that we have now. The variance in data is certainly a concern, but the free/inexpensive nature of the app itself was intended to gather as much as possible (anonymity being absolutely key, given the subject) and start looking at data points vs. positive ratings to find the correlation.

At the very least, that sort of data would be incredibly value for academia/research etc.

Shoot me an email (see HN profile/website) if you ever think about reviving the idea. I would be very interested to colloborate.

Reply on news.ycombinator.com