I've been kicking this idea around the past month or so.
I'm contemplating developing a machine learning, neural network to process
against NCAA basketball statistical data. The data is available in CSV file format, also
compatible with Excel.
There's lots of other data that I want to include that takes advantage of machine learning.
Nothing is ever going to be exact, but I believe this might be the best way of accomplishing something darn good.
I'm willing to do all the software programming, plus data neural networks is much easier than the stuff I'm working on lol.
I'ld like to collaborate with others that wouldn't mind doing the work of gathering
the data and collaborating on what goes into the neural network. You don't need to
understand what a neural network is, just think and use statistical data, trends etc
that would give a great insight to rankings. Rankings being broken down into a lot
of metrics.
This type of system is so much better than the flat files and basic guessing, especially
at the beginning of the seasons. Requires more data and more work, but a better outcome.
It's going to require a lot of different datasets. Some easy to get, some not so
easy.
If you're interested or have ideas on stats that should go into this, let me know!
I'm contemplating developing a machine learning, neural network to process
against NCAA basketball statistical data. The data is available in CSV file format, also
compatible with Excel.
There's lots of other data that I want to include that takes advantage of machine learning.
Nothing is ever going to be exact, but I believe this might be the best way of accomplishing something darn good.
I'm willing to do all the software programming, plus data neural networks is much easier than the stuff I'm working on lol.
I'ld like to collaborate with others that wouldn't mind doing the work of gathering
the data and collaborating on what goes into the neural network. You don't need to
understand what a neural network is, just think and use statistical data, trends etc
that would give a great insight to rankings. Rankings being broken down into a lot
of metrics.
This type of system is so much better than the flat files and basic guessing, especially
at the beginning of the seasons. Requires more data and more work, but a better outcome.
It's going to require a lot of different datasets. Some easy to get, some not so
easy.
If you're interested or have ideas on stats that should go into this, let me know!
