@Article{          janssen.ea2017-finding,
    author       = {Janssen, Berit and van Kranenburg, Peter and Volk, Anja},
    year         = {2017},
    title        = {Finding Occurrences of Melodic Segments in Folk Songs
                   Employing Symbolic Similarity Measures},
    abstract     = {Much research has been devoted to the classification of
                   folk songs, revealing that variants are recognised based
                   on salient melodic segments, such as phrases and motifs,
                   while other musical material in a melody might vary
                   considerably. In order to judge similarity of melodies on
                   the level of melodic segments, a successful similarity
                   measure is needed which will allow finding occurrences of
                   melodic segments in folk songs reliably. The present study
                   compares several such similarity measures from different
                   music research domains: correlation distance, city block
                   distance, Euclidean distance, local align-ment, wavelet
                   transform and structure induction. We evaluate the
                   measures against annotations of phrase occurrences in a
                   corpus of Dutch folk songs, observing whether the
                   mea-sures detect annotated occurrences at the correct
                   positions. Moreover, we investigate the influence of music
                   represen-tation on the success of the various measures,
                   and analyse the robustness of the most successful measures
                   over subsets of the data. Our results reveal that
                   structure induction is a promising approach, but that
                   local alignment and city block distance perform even
                   better when applied to adjusted music representations.
                   These three methods can be combined to find occurrences
                   with increased precision.},
    doi          = {10.1080/09298215.2017.1316292},
    issn         = {0929-8215},
    journal      = {Journal of New Music Research},
    keywords     = {music analysis with computers,music
                   similarity,occurrences,pattern
                   matching,segments,symbolic},
    mendeley-tags= {music analysis with computers},
    month        = {apr},
    number       = {2},
    pages        = {118--134},
    url          = {https://www.tandfonline.com/doi/full/10.1080/09298215.2017.1316292},
    volume       = {46}
}

{"_id":"iMYSJFjoyZaN4TTzq","bibbaseid":"janssen-vankranenburg-volk-findingoccurrencesofmelodicsegmentsinfolksongsemployingsymbolicsimilaritymeasures-2017","authorIDs":[],"author_short":["Janssen, B.","van Kranenburg, P.","Volk, A."],"bibdata":{"bibtype":"article","type":"article","author":[{"propositions":[],"lastnames":["Janssen"],"firstnames":["Berit"],"suffixes":[]},{"propositions":["van"],"lastnames":["Kranenburg"],"firstnames":["Peter"],"suffixes":[]},{"propositions":[],"lastnames":["Volk"],"firstnames":["Anja"],"suffixes":[]}],"year":"2017","title":"Finding Occurrences of Melodic Segments in Folk Songs Employing Symbolic Similarity Measures","abstract":"Much research has been devoted to the classification of folk songs, revealing that variants are recognised based on salient melodic segments, such as phrases and motifs, while other musical material in a melody might vary considerably. In order to judge similarity of melodies on the level of melodic segments, a successful similarity measure is needed which will allow finding occurrences of melodic segments in folk songs reliably. The present study compares several such similarity measures from different music research domains: correlation distance, city block distance, Euclidean distance, local align-ment, wavelet transform and structure induction. We evaluate the measures against annotations of phrase occurrences in a corpus of Dutch folk songs, observing whether the mea-sures detect annotated occurrences at the correct positions. Moreover, we investigate the influence of music represen-tation on the success of the various measures, and analyse the robustness of the most successful measures over subsets of the data. Our results reveal that structure induction is a promising approach, but that local alignment and city block distance perform even better when applied to adjusted music representations. These three methods can be combined to find occurrences with increased precision.","doi":"10.1080/09298215.2017.1316292","issn":"0929-8215","journal":"Journal of New Music Research","keywords":"music analysis with computers,music similarity,occurrences,pattern matching,segments,symbolic","mendeley-tags":"music analysis with computers","month":"apr","number":"2","pages":"118–134","url":"https://www.tandfonline.com/doi/full/10.1080/09298215.2017.1316292","volume":"46","bibtex":"@Article{ janssen.ea2017-finding,\n author = {Janssen, Berit and van Kranenburg, Peter and Volk, Anja},\n year = {2017},\n title = {Finding Occurrences of Melodic Segments in Folk Songs\n Employing Symbolic Similarity Measures},\n abstract = {Much research has been devoted to the classification of\n folk songs, revealing that variants are recognised based\n on salient melodic segments, such as phrases and motifs,\n while other musical material in a melody might vary\n considerably. In order to judge similarity of melodies on\n the level of melodic segments, a successful similarity\n measure is needed which will allow finding occurrences of\n melodic segments in folk songs reliably. The present study\n compares several such similarity measures from different\n music research domains: correlation distance, city block\n distance, Euclidean distance, local align-ment, wavelet\n transform and structure induction. We evaluate the\n measures against annotations of phrase occurrences in a\n corpus of Dutch folk songs, observing whether the\n mea-sures detect annotated occurrences at the correct\n positions. Moreover, we investigate the influence of music\n represen-tation on the success of the various measures,\n and analyse the robustness of the most successful measures\n over subsets of the data. Our results reveal that\n structure induction is a promising approach, but that\n local alignment and city block distance perform even\n better when applied to adjusted music representations.\n These three methods can be combined to find occurrences\n with increased precision.},\n doi = {10.1080/09298215.2017.1316292},\n issn = {0929-8215},\n journal = {Journal of New Music Research},\n keywords = {music analysis with computers,music\n similarity,occurrences,pattern\n matching,segments,symbolic},\n mendeley-tags= {music analysis with computers},\n month = {apr},\n number = {2},\n pages = {118--134},\n url = {https://www.tandfonline.com/doi/full/10.1080/09298215.2017.1316292},\n volume = {46}\n}\n\n","author_short":["Janssen, B.","van Kranenburg, P.","Volk, A."],"key":"janssen.ea2017-finding","id":"janssen.ea2017-finding","bibbaseid":"janssen-vankranenburg-volk-findingoccurrencesofmelodicsegmentsinfolksongsemployingsymbolicsimilaritymeasures-2017","role":"author","urls":{"Paper":"https://www.tandfonline.com/doi/full/10.1080/09298215.2017.1316292"},"keyword":["music analysis with computers","music similarity","occurrences","pattern matching","segments","symbolic"],"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"article","biburl":"https://hmb.sampaio.me/bibliografia.bib.txt","creationDate":"2019-04-22T13:04:36.099Z","downloads":0,"keywords":["music analysis with computers","music similarity","occurrences","pattern matching","segments","symbolic"],"search_terms":["finding","occurrences","melodic","segments","folk","songs","employing","symbolic","similarity","measures","janssen","van kranenburg","volk"],"title":"Finding Occurrences of Melodic Segments in Folk Songs Employing Symbolic Similarity Measures","year":2017,"dataSources":["n6MFY2CscQLDpJ7nT","RFLDZw5KyJdadDXDm"]}