Tim Thompson
Librarian for Applied Metadata Research at Yale University Library
Text Embeddings for Entity Resolution and Name Disambiguation in the Library Catalog
Tim Thompson
Librarian for Applied Metadata Research
Yale University Library
timothy.thompson@yale.edu
www.linkedin.com/in/timathompson
Yale Library AI Community of Interest
October 9, 2024
Attribution & Acknowledgments
- License
- Creative Commons Attribution 4.0 International (CC-BY).
- Credits
-
- Catherine Kwon (Yale, B.S. in Statistics & Data Science and Art)
- Data Chemist
- Gavin Mendel-Gleason
- Maren van Otterdijk
- Daniel Lovins
- Marty Kurth
- IT Steering Committee
- Library IT
- Cultural Heritage IT (CHIT) Initiative
Overview
- Issue
- Problem statement and key concepts
- Action
- Project scope and evolution
- Impact
-
Initial results, challenges, potential next steps
[Photograph of Beinecke Construction]. 0AD. https://collections.library.yale.edu/catalog/2037090.
Work in Progress
Authority Control
Who’s really in control here?
- Users depend on consistent access points to find what they’re looking for.
- Access points for people, groups, subjects (and to some extent works) are meant to be controlled and “authorized” against master files.
- In practice, authority control is messy because data is denormalized and entities are identified by strings rather than unique identifiers.
Authority Control in the Catalog
Catalog card taken from the digital version of the book Documentation made easy.
This Heading Is Valid?
Will the Real Schubert Please Stand Up?
- The lack of consistent, comprehensive authority control in the catalog is a form of technical debt that impacts both users and staff.
- The migration from Voyager to Alma gives us an opportunity to address the problem.
- Alma supports integrated authority control processes, but we need to improve the quality of our data to utilize this functionality effectively.
Problem Statement
Entity Resolution
With some help from text embeddings?
1
2022–2023
Student employees (CHIT and library funded) used an Excel plugin to perform manual entity resolution and disambiguation for common names in the catalog
3
November 2023
Proposal to Marty Kurth and ITSC
5
September–October 2024
- Hiring students
- Following up with DataChemist on bug fixes and optimizations
- Carrying out further testing
2
2023
TerminusDB (graph database) and VectorLink: research and testing
4
February–March 2024
Consultant project with DataChemist (TerminusDB and VectorLink developers)
Initial Steps
Consultant Project
Key Concepts
Text embeddings, vector databases, HNSW, ANN...
-
Embedding models transform text into a numeric representation.
-
Given a word or string of words, embedding models output a high-dimensional vector that captures the text’s semantic meaning, syntactic structure, and contextual nuance.
Text Embeddings
Credit: Catherine Kwon
[0.004, -0.005, ..., -0.017, 0.005]
“Today is a great day!”
string of text
high-dimensional vector
Text Embeddings
EMBEDDING
MODEL
Credit: Catherine Kwon
Credit: Catherine Kwon
Dimensionality Reduction
Principal Component Analysis (PCA)
Ground Transportation
Credit: Catherine Kwon
- Once you have a set of text embeddings, what do you do with them?
- Store them in a specialized index!
- Vector databases use indexing algorithms such as Hierarchical Navigable Small World (HNSW) graphs for Approximate Nearest Neighbor (ANN) search.
- As search engines, vector databases make it possible to return results based on user intent, rather than simple keyword matching.
Vector Databases
Credit: Catherine Kwon
Project Outcomes
Pipeline
- Export from Voyager
1
Extract
- Transform to BIBFRAME
- Generate embeddings strings
2
Transform
- Embed using OpenAI
- Index vectors in VectorLink
3
Load
Status
- Currently focused on people as contributors (vs. subjects).
- Testing with the curated benchmark dataset produced 2022–2023.
- 1,274 entries across 54 name/identity clusters.
Embedding Strings
Contributor: Schubert, Franz
Title: The art of photography: instructions in the art of producing photographic pictures in any color, and on any material : for the use of beginners : and also of persons who have already attained some proficiency in the art : and of engravers on copper, stone, wood, etc.
Subjects: Photography
Embedding Objects
{
"op": "Inserted",
"string": "Contributor: Schubert, Franz\n Title: The art of photography: instructions in the art of producing photographic pictures in any color, and on any material : for the use of beginners : and also of persons who have already attained some proficiency in the art : and of engravers on copper, stone, wood, etc.\n Subjects: Photography\n",
"marcKey": "7001 $aSchubert, Franz.",
"person": "Schubert, Franz",
"roles": "Contributor",
"title": "The art of photography: instructions in the art of producing photographic pictures in any color, and on any material : for the use of beginners : and also of persons who have already attained some proficiency in the art : and of engravers on copper, stone, wood, etc.",
"variant_titles": null,
"hub_title": null,
"subjects": "Photography",
"genres": null,
"record": "14703468",
"id": "14703468#Agent700-23"
}
VectorLink
- Once embeddings are indexed, VectorLink calculates the “neighborhood” of each node based on a similarity threshold.
- The results can be converted to an edge list for graph visualization.
“Schubert, Franz”
- 88 records
- All 4 identities correctly clustered
- 4 unconnected nodes
- ~95% accuracy
- Threshold: 0.16
1
2
3
4
“Thayer, Nathaniel”
- 26 records
- 3 (4?) identities correctly clustered
- Some fragmentation of the main cluster
- Threshold: 0.16
1
2
3
4?
1a
1b
- Every person is different! Difficult to set a uniform threshold for similarity.
- Trying to model a person’s identity (or identities) through their works is at best an approximation.
- Difficult to “curate” benchmark data and embedding strings.
-
Pipeline is not yet optimized for production.
- Embeddings are cheap (~$15.00 to embed 13 million strings using text-embedding-3-small)
- Computation and storage are expensive (7.5G of embedding strings equals 75G of vectors).
Challenges and Limitations
Redundant Data
Contributor: Halleur, G. C. Hermann
Title: The art of photography: instructions in the art of producing photographic pictures in any color, and on any material : for the use of beginners : and also of persons who have already attained some proficiency in the art : and of engravers on copper, stone, wood, etc.
Subjects: Photography
Contributor: Schubert, Franz
Title: The art of photography: instructions in the art of producing photographic pictures in any color, and on any material : for the use of beginners : and also of persons who have already attained some proficiency in the art : and of engravers on copper, stone, wood, etc.
Subjects: Photography
Library of Congress Names?
Name: Schubert, Franz, 1876-
Variant names: Schubert, Franz, b. 1876
Sources: Grundzüge der Pastoraltheologie, 1922 (Dr. Franz Schubert, o. ö. Professor an der Universität Breslau); DNB in VIAF, Oct. 7, 2011 (hdg.: Schubert, Franz, 1876-; German theologian, professor of pastoral theology); Deutsche Biographie, viewed 28 September 2022 (Franz Schubert; born in Bistrai-Bielitz (Austrian Silesia) in 1876, died in Breslau in 1937; Catholic theologian specializing in pastoral theology)
Contributor: Schubert, Franz
Title: Liturgische Zeitschrift
Subjects: Catholic Church--Liturgy--Periodicals.
Similarity: 0.23 😿
Bonus!
Simple RAG demo
Thank you!
Tim Thompson
Text Embeddings for Entity Resolution and Name Disambiguation in the Catalog
By Tim Thompson
Text Embeddings for Entity Resolution and Name Disambiguation in the Catalog
Discover innovative approaches to entity resolution and name disambiguation using text embeddings with library catalog data. Explore the intriguing challenges and initial findings of this ongoing project, and see how technology can enhance authority control!
