Implementing Genre and Medium of Performance Algorithm

Slide Note
Embed
Share

This project outlines the process of implementing an algorithm to categorize music records based on genre and medium of performance. Starting with a test run, the algorithm processes music MARC records, identifies headings, and generates output files. The initial evaluation reveals changes made, potential cleanup requirements, and problems encountered. The assessment phase involves manual verification, heading analysis, and creating spreadsheets to review and analyze generated headings. The project aims to streamline the categorization process for a local music catalog.


Uploaded on Sep 29, 2024 | 0 Views


Download Presentation

Please find below an Image/Link to download the presentation.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.

E N D

Presentation Transcript


  1. Implementing the Genre and Medium of Performance Algorithm in a Local Catalog Jeff Lyon jeff_lyon@byu.edu Brigham Young University Greg Reeve greg_reeve@byu.edu

  2. Process Roadmap Initial test run Assessment phase Local music cataloging procedures Record pull Pre-processing Run algorithm Post-processing Load changes Instrumentation search

  3. Initial test run Extract all music MARC records Remove local fields Install the MusicFormGenre tool Run the tool on the extracted records Verify output files Load into test ILS server

  4. Algorithm output files MusicFormGenre.20200204.212205.NothingHappened.mrc MusicFormGenre.20200204.212205.NothingHappened.txt MusicFormGenre.20200204.212205.RecordsChanged.Message.mrc MusicFormGenre.20200204.212205.RecordsChanged.Message.txt MusicFormGenre.20200204.212205.RecordsChanged.OK.DeleteReason.mrc MusicFormGenre.20200204.212205.RecordsChanged.OK.DeleteReason.txt MusicFormGenre.20200204.212205.RecordsChanged.OK.mrc MusicFormGenre.20200204.212205.RecordsChanged.OK.txt MusicFormGenre.20200204.212205.Report.txt

  5. Evaluating initial run output files RecordsChanged.OK RecordsChanged.Message RecordsChanged.OK.DeleteReason NothingHappened Problems 212,544 76,126 5,395 51,694 0

  6. Assessment phase Manual record verification Heading analysis 650 heading report/creating the spreadsheet Actual changes and potential future cleanup

  7. Assessment phase: Heading analysis Extract 650 and 655 headings Generate a unique list of headings with usage count Generate dummy MARC records for each heading Run dummy MARC records through the musicformgenre tool Programmatically analyze the output for each heading Generate a spreadsheet with each heading and usage count combined with basic checks about the heading output Review the top 200 headings and their output

  8. Assessment phase: Heading analysis 22,899 8,149 8,008 6,608 6,137 6,036 5,910 5,249 4,804 4,152 4,119 3,843 Songs with piano Operas--Excerpts Symphonies Organ music Orchestral music Christmas music Sacred music Harp music Jazz Songs (High voice) with piano Motion picture music--Excerpts Sonatas (Piano) 80,896 32,062 27,831 6,932 Long-playing records 78 rpm records Compact discs 45 rpm records

  9. Assessment phase: Heading analysis Required cleanup Potential future cleanup Flip Lost demographic Misc Chamber music Nothing produced Canons, fugues, etc. No 382 No genre No useful genre Data errors

  10. Assessment phase: Stats

  11. Assessment phase: Stats

  12. Assessment phase: Stats

  13. Local music cataloging rule set

  14. Pre-run steps Extract music MARC records Pre-processing on extracted records Problematic metadata Viola and music|vScores and parts. Crashed the program

  15. Run algorithm

  16. Post-processing & load changes Add 583 tag Remediation based on our heading analysis Coordinate with Library IT to load records Identify modified records Track the category of change made

  17. Code repository https://tinyurl.com/byugenre

  18. Instrumentation search

  19. Lessons learned & future steps

  20. Implementing the Genre and Medium of Performance Algorithm in a Local Catalog https://tinyurl.com/byugenre Jeff Lyon jeff_lyon@byu.edu Brigham Young University Greg Reeve greg_reeve@byu.edu

Related