Composer Attribution of Renaissance Motets (Iberian Polyphony around 1500): MIDIs and Extracted Features
Bibliographic record
Abstract
This distribution includes MIDI files used in the experiments described in the "Composer Attribution of Renaissance Motets: A Case Study Using Statistical Features and Machine Learning" chapter of the book The Anatomy of Iberian Polyphony around 1500. Multi-part motets have been separated out into separate MIDI files. We edited the files for consistency, which included adjusting elements that could interfere with encoding rhythm consistently (e.g. standardized rhythmic value rates, eliminating fermatas, etc.). The MIDI files in the nonIb_noMB group were taken from the Josquin Research Project (JRP), who have kindly granted us permission to re-publish them here with our modifications. These are redistributed with a "CC BY-SA 4.0" license: https://github.com/josquin-research-project/jrp-scores/blob/master/LICENSE.txt. The Iberian MIDI files were produced from editions created by The Anatomy of Late 15th- and Early 16th-Century Iberian Polyphonic Music project (https://iberianpolyphonicmusic.wordpress.com), and are included here after our modifications, with permission. We digitised the nonIb_MBonly MIDI files ourselves using Sibelius. All these files are distributed under a "CC BY-SA 4.0" license" license (https://creativecommons.org/licenses/by-sa/4.0/). The included "Catalogue.pdf" file outlines the contents of this corpus in its entirety. This distribution also includes the "Renaissance-safe" features extracted using jSymbolic 2.2 (http://jmir.sourceforge.net) from the MIDI encodings included here. Details on all the features extracted with the software are available in the jSymbolic manual (http://jmir.sourceforge.net/manuals/jSymbolic_manual/home.html). These are presented as follows: - FeatureDefinitions.xml: Descriptions of all extracted features, encoded in ACE XML 1.0, as output directly by jSymbolic. This file does not include any feature values (these are found in the FeatureValues.xml file). - FeatureValues.xml: Extracted feature values, encoded in ACE XML 1.0, as output directly by jSymbolic. The features are described in the FeatureDefinitions.xml file. Class values are implied by the folder containing each MIDI file. - FeatureValues_BasicCSV: Extracted feature values encoded in a CSV file, as output directly by jSymbolic (with complete file paths truncated). Class values are implied by the folder containing each MIDI file. - FeatureValues_HumanReadable.xlsx: Extracted features formatted into a human-readable Microsoft Excel file. Group averages and standard deviations have been added to the bottom, and the "Length_In_Breves" feature is added in a column at the right (it is included separately because it is not calculated directly by jSymbolic 2.2). - FeatureValues_WekaReady.csv: Extracted feature values encoded in a CSV file in a format readable by Weka (https://www.cs.waikato.ac.nz/ml/weka/). Class values have been added in a column on the right, and file paths have been removed, as required by Weka.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.009 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.003 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.119 | 0.077 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".