#4 Musical Heritage Knowledge Extraction from text
WP4 develops methods and tools for 1) creating plurilingual corpora focused on musical heritage, 2) extracting (automatically) relevant knowledge from text such as people, places, time, music works, experiences, stories, events, etc. about musical heritage, 3) evaluating knowledge extraction tools against the manually curated corpora. The developed methodologies and tools will enable the processing of multilingual textual sources, which express diverse perspectives at different times: linguistic, historical, cultural, political, etc.
Related components:
- Corpus
- Repository
- WebApplication
- Dataset
- SPARQLEndpoint
- KnowledgeGraph
- CLITool
- DockerImageContainer
- Project
- Documentation
- Software
- Curation of documentary evidence, experiments with LED/GPT-4
- MEETUPS Corpus collection
- MEETUPS - Identification of temporal knowledge
- MEETUPS preparation - data cleaning
- MEETUPS - Historical meetups identification
- MEETUPS Identification of themes
- MEETUPS identification of people and places
- MEETUPS - Identification of temporal knowledge
- Ontology
- Lexicon
- Story
- Hearing Music
- Sources Cross-Analysis
- Music Historian
- Restoration and Sound Practices
- Musical Social Network
- Conflicting theoretical interpretations
- Conflicting analytical annotations
- Exploration mode
- Shuffle mode
- DiscourseAnalysis
- Terminology
- Persona