In Search of Zora/When Metadata Isn’t Enough: Rescuing the Experiences of Black Women Through Statistical Modeling

Nicole M. Brown, Ruby Mendenhall, Michael Black, Mark Van Moer, Karen Flynn, Malaika McKee, Assata Zerai, Ismini Lourentzou, Cheng Xiang Zhai

Research output: Contribution to journalArticle

Abstract

This study used statistical topic modeling to examine 800,000 documents within HathiTrust and JSTOR databases to identify the kinds of discourses in books, poetry, newspapers, and journals related to African American women. We examined a range of conversations that emerged, between 1746 and 2014, revealing insights about, and from African American women. We identified a metadata revision methodology that served to rescue 150 documents for or about Black women that were either not previously cataloged or cataloged in such a way that Black women's experiences are either lost or erased. This project’s use of computation is unique in that it allows for the quantitative surveying of such a large dataset while charting a qualitative assessment to determine if and how texts capture the experiences of African American women. Using a technique called ‘intermediate reading’, texts are verified for their applicability. This strategy of search, recognition, rescue and recovery (SeRRR) may aid curators of information in making Black women’s voices more accessible within the digitized record. The SeRRR strategy will allow scholars to use a form of ‘call and response’ with metadata to understand the lived (and death) experiences of Black women as Alice Walker did during her search for Zora Neal Hurston.

Original languageEnglish (US)
Pages (from-to)141-162
Number of pages22
JournalJournal of Library Metadata
Volume19
Issue number3-4
DOIs
StatePublished - Oct 8 2019

Keywords

  • African American women
  • metadata
  • methodology
  • topic modeling

ASJC Scopus subject areas

  • Library and Information Sciences

Fingerprint Dive into the research topics of 'In Search of Zora/When Metadata Isn’t Enough: Rescuing the Experiences of Black Women Through Statistical Modeling'. Together they form a unique fingerprint.

  • Cite this