July 28, 2017

Artificial Intelligence

This November AquaticLog will be turning 5 years old. The industry has seen a lot of positive changes in the way aquarists look after their reefs and there is now more knowledge and tools at their disposal.

One important aspect of reef keeping is to have a handle on water parameters. While some can be meticulous about every single parameter they can test for, others practice a much more lax approach and many are in between. No matter what your routine is like, you are guaranteed to keep keep an eye on your temperature, salinity and alkalinity if you want to be successful at keeping corals.



The combination of parameter ranges for a reef tank is as unique as your fingertips and may fluctuate daily. You can keep a low nutrient system such as Zeovit or can be completely off the charts for Nitrates and Phosphates and yet have a reef as amazing as Richard Ross' aquarium. Whether you are new to the hobby or are an old salt, you will target some parameter ranges.

We all know general guidelines and acceptable ranges, they are published in every book or forum article. You can try and mimic someone else's parameters. But what if there was a way to get suggestions on a specific set of parameter values that are unique to your aquarium?

Have you ever heard of Machine Learning? This Artificial Intelligence technology is what Amazon and Netflix would use to recommend you books and movies that are tailored to your taste. AquaticLog is the reef industry first to use Machine Learning to learn from 840,000+ measurements logged on the site and recommend a precise set of parameters that should take your reef to the state of equilibrium bliss.

The algorithm takes your current parameters (you can specify which parameters to use) and will recommend the adjustments or will confirm that they are in a good state. If you prefer to keep a high alkalinity system it will suggest different parameters then for a natural sea water aquarium. If you prefer to have low nitrates it will suggest a value for phosphates that is appropriate. If your system runs cooler than most it will find matching parameters as well.



This advanced algorithm would not be possible without heavy math, big data and thousands of successful aquariums logged using AquaticLog. We have been running this algorithm in beta for a few weeks and results have been eerily good.

We are excited to hear what your experience will be with this revolutionary new feature.

Keep on reefing.

statru ‐ AWESOME!!! <*}}}><
pjvalencia ‐ Love it. keep it up.