I often talk to people interested in what Thingworx Analytics can offer, and generally those people fall into two different camps... Some don't know what ML / AI is and are interested in better understanding of what kind of things can be done with it and Thingworx Analytics. And some do know what ML / AI is from the inside and they want to understand how Thingworx Analytics works from the inside and how to map TW Analytics terms and concepts to "standard" methods and algorithms. I myself know the math / algorithms behind ML, so talking to the latter group is kinda easier for me, but I still always on the lookout for more details.
So recently I was searching for more materials to help myself to communicate the value of Thingworx Analytics, as well as better understand some technical stuff, and found a few nice Thingworx Analytics videos (some are pretty outdated, but still really helpful), which I found to be interesting, so wanted to share with the community.
The first one is about Thingworx Analytics Signals. It fits great for the guys who want to know how Analytics works "under the hood" and explains what's the logic behind Signals and how to interpret its results. I could kinda guess / understand how it works from the algorithmic point of view, but the video clarifies some details and has a good explanation of the results' interpretation. https://www.youtube.com/watch?v=wfupQQYczWo Would be great to have something similar on profiles too, but I couldn't find.
The other video - "Predictive Maintenance with Thingworx 101" is from LiveWorx 2018 and had more exposure, but for those who haven't seen it, I think it's worth it.
It does a great job at explaining the bigger picture and helps with understanding of concepts behind ML, what a ML project looks like and what kind problems could be solved using Thingworx Analytics. https://www.youtube.com/watch?v=1o02FJFd6ss