I came across some good advice on maintenance of Machine Learning algorithms. The short version, Less is More when you are deciding what data to feed into your algorithm.
It also reminded me of some systems issues Jaron Lanier pointed out in a recent interview.
- Over time will your model be primarily processing its own predictions?The Netflix movie recommendation engine has narrowed its available inputs because most users only choose between the top predictions.
- Will your prediction model cannibalize its source of data?Machine learning powered language translation is automating the work of many human translators, whose previous work provides the input needed by the machine learning algorithms.