information retrieval evaluation music information retrieval mirex reliability mir trec sigir validity statistical significance music similarity crowdsourcing randomization music bootstrap permutation wilcoxon information extraction t-test phd symbolic melodic similarity cranfield statistics ir melodic similarity kendall correlation test collection ismir2011 audio relevance similarity interpolation relevance judgment experiment education meta-evaluation admire efficiency audio music similarity ismir2012 ismir effectiveness science sigir2013 ictir2017 ties sigir2018 information retrieval evaluation simulation average precision stochastic copula ismir2018 hypothesis testing research student doctorate sigir2019 statistical testing type i type ii type iii chroma mp3 mfcc robustness sign test ground truth alignment annotation pattern melody extraction ieg grammar named entity recognition user satisfaction dcg sigir2016 estimation ismir2016
See more