Data-Driven Relevance Judgments for Ranking Evaluation

In this repository you will find the code and data used to execute the experimental process (designed for R) of the paper "Data-Driven Relevance Judgments for Ranking Evaluations".

This article was written by Nuno Moniz, Luís Torgo and João Vinagre. It was uploaded to arXiv.

Figure1.R : R Code to generate Figure 1
GoD_phi.R : R Code to interpolate relevance according to the magnitude of their divergence
SyntheticExps.R : R Code to replicate synthetic experiments described in the paper and to generate Figure 2
RealWorldExps.R : R Code to replicate real world experiments described in the paper and to generate Figure 3

CSV's contain the data used to replicate the real world experiments (Figure 3)