Closed pacastillo closed 8 years ago
The re-opener is here. :D
I think the figure should specify a bit deeper the methodology. Thus I propose to include in each block something like:
DATA COLLECTION: 6000 real books - 50 features
PREPROCESSING: Select pre-sales data, feature refinement
FEATURE SELECTION: SOM, RelFAttEval, CFSubEval
FORECASTING: Regression techniques (kNN, M5P, RF, LR, SVM, MLP, ELM)
VALIDATION: Decision Trees/Regression Models
What do you think?
Ok :)
related to R3(comment 3)