Towards Musicologist-Driven Mining of Handwritten Scores

• Published in: IEEE intelligent systems Vol. 33; no. 4; pp. 24 - 34
• Main Authors: Niitsuma, Masahiro, Tomita, Yo, Yan, Wei Qi, Bell, David
• Format: Journal Article
• Language: English
• Published: Los Alamitos IEEE 01-07-2018; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Barometers; Chronology; Data mining; domain-driven data mining; Feature extraction; Handwriting; historical musicology; Intelligent systems; Knowledge discovery; music informatics; music information retrieval; Musicology; Object recognition; optical music recognition; Radio frequency; Support vector machines
• ISSN: 1541-1672; 1941-1294
• DOI: 10.1109/MIS.2018.111144115

Historical musicologists have been seeking objective and powerful techniques to collect, analyze, and verify their findings for many decades. The aim of this study was to show the importance of such domain-specific problems to achieve actionable knowledge discovery in the real world. Our focus is on finding evidence for the chronological ordering of J.S. Bachs manuscripts, by proposing a musicologist-driven mining method for extracting quantitative information from early music manuscripts. Bachs C-clefs were extracted from a wide range of manuscripts under the direction of domain experts, and with these, the classification of C-clefs was conducted. The proposed methods were evaluated on a dataset containing over 1000 clefs extracted from J.S. Bachs manuscripts. The results show more than 70% accuracy for dating J.S. Bachs manuscripts. Dating of Bachs lost manuscripts was quantitatively hypothesized, providing a rough barometer to be combined with other evidence to evaluate musicologists hypotheses, and the usability of this domain-driven approach is demonstrated.