Recommendations and how to measure the ROI with some metrics ?
From the post:
We talked a lot about recommender systems, specially discussing the techniques and algorithms used to build and evaluate algorithmically those systems. But let’s discuss now how can we measure in quantitative terms how a social network or an on-line store can measure the return of investment (ROI) of a given recommendation.
The metrics used in recommender systems
We talk a lot about F1-measure, Accuracy, Precision, Recall, AUC, those buzzwords widely known by the machine learning researchers and data mining specialists. But do you know what is CTR, LOC, CER or TPR ? Let’s explain more about those metrics and how they can evaluate the quantitative benefits of a given recommendation.
Would you feel more comfortable if I said identification instead of recommendation?
Consider it done.
After all, a “recommendation” is some actor making a statement about identified subject. Run of the mill stuff for a topic map.
The ROI question is whether there is some benefit to that statement + identification?
Assuming you are using a topic map or similar measures to track the source of a recommendation, you could begin to attach ROI to particular sources of recommendation.