Machine learning to predict yield

I am wanting to use a combination of numeric and categorical data to predict numeric data using matlab/machine learning. I'm wondering which application would do this. For example for a vineyard, I have weather data - which is the same for the vineyard and am wanting to use this to predict yield for different blocks of fruit. The only input data that I can say is different between these blocks is the soil type of the block. Which application could incorporate both this numeric and categorical data?

Answer 1

[Kartik Mittal]

I think the right approach would be to convert the categorical variable into a numeric one by encoding it in numeric levels and probably it would work better if you normalize your attributes after that. You can refer to material related to dummy coding for categorical variables. After that you can leverage different algorithms in MATLAB's Statistics toolbox for predictive modelling.

[Wendy Cameron]

Thanks very much, I'll follow that up.

Wendy