Review of model results thanks to the InterpretML library After all these operations we can finally look at the model and try to draw some conclusions about the characteristics of our assortment. set_visualize_provider InlineProvider ebm_global = ebm.explain_global show ebm_global Summary of the significance of model features The basic view is a summary of the features that were most important for the model when predicting the class.
It allows you to assess which features are worth looking at. The distribution of the significance of the features is also important in our case none of the features dominates the others. product_category The most important feature from the model's point of view is product_category. It clearly indicates which product categories are Taiwan WhatsApp Number List more desirable by customers and which are the opposite. Score is a relative value significant within the feature of the current model and should be treated as such. Negative values increase the probability of belonging to cluster positive values to cluster and values close to are neutral for the decision. Density is how much data of that type value is in the set keep this in mind especially if it's very low. n_unit_of_measure An interesting example for analysis is the n_unit_of_measure feature which was ranked in the top in terms of significance. At values > . the model records a jump in probability towards cluster this cluster is the more valuable in our case.
Perhaps it is worth considering changing the packaging grammage for some products? promo_yn Another case worth analyzing is the products covered by the promotion in the store. For the model such a category drastically increases the probability of assigning a product to cluster low value . Perhaps this means that our marketing activities are ineffective and it is worth considering changes in this area . The model also analyzes the relationships of individual features in the context of changes in probability within clusters. However in our case these relations are small in global terms.