Web Archives Metadata Generation with GPT-4o: Challenges and Insights.
Current metadata creation for web archives is time consuming and costly due to reliance on human effort. This paper explores the use of GPT-4o for metadata generation within the Web Archive Singapore, focusing on scalability, efficiency, and cost effectiveness. We processed 112 Web ARChive (WARC) fi...
| Publicado en: | Information Technology & Libraries Vol. 44; no. 2; pp. 1 - 17 |
|---|---|
| Autores principales: | , , , |
| Formato: | research tables/charts Journal Article |
| Publicado: |
American Library Association
Jun2025
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=186511082&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 186511082 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 07309295 ITL jtl: Information Technology & Libraries issn: 07309295 maglogo: N pubinfo: dt: Jun2025 vid: 44 iid: 2 pid: 55 pub: American Library Association place: Chicago, Illinois artinfo: ui: 186511082 186511082 186511082 10.5860/ital.v44i2.17305 186511082 ppf: 1 ppct: 16 formats: fmt: @attributes: type: P tig: atl: Web Archives Metadata Generation with GPT-4o: Challenges and Insights. aug: au: Nair, Ashwin Zhen Rong Goh Tianrui Liu Yongping Huang, Abigail affil: Manager/Librarian (Systems), Resource Discovery Management, National Library Board, Singapore sug: subj: Artificial Intelligence, Generative Utilization Natural Language Processing Methods Metadata World Wide Web Singapore Archives Cost Effectiveness Analysis Human Singapore Biomedical Engineering Algorithms Information Storage Information Retrieval Resource Databases Internet Cost Savings Libraries, Electronic Data Management Methods Motivation Automation Information Technology Descriptive Statistics Comparative Studies Software Computer Hardware Nonparametric Statistics Privacy and Confidentiality ab: Current metadata creation for web archives is time consuming and costly due to reliance on human effort. This paper explores the use of GPT-4o for metadata generation within the Web Archive Singapore, focusing on scalability, efficiency, and cost effectiveness. We processed 112 Web ARChive (WARC) files using data reduction techniques, achieving a notable 99.9% reduction in metadata generation costs. By prompt engineering, we generated titles and abstracts, which were evaluated both intrinsically using Levenshtein distance and BERTScore, and extrinsically with human cataloguers using McNemar's test. Results indicate that while our method offers significant cost savings and efficiency gains, human curated metadata maintains an edge in quality. The study identifies key challenges including content inaccuracies, hallucinations, and translation issues, suggesting that large language models (LLMs) should serve as complements rather than replacements for human cataloguers. Future work will focus on refining prompts, improving content filtering, and addressing privacy concerns through experimentation with smaller models. This research advances the integration of LLMs in web archiving, offering valuable insights into their current capabilities and outlining directions for future enhancements. The code is available at https://github.com/masamune-prog/warc2summary for further development and use by institutions facing similar challenges. pubtype: Academic Journal doctype: research tables/charts Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|