Undressing Fashion Metadata: Ryerson University Fashion Research Collection

Abstract

The purpose of this poster is to provide insight into the processes involved in making a unique fashion research and teaching collection discoverable in an online environment at Ryerson University. The online collection will provide a means for the users to identify what artifacts are available for research purposes and facilitate teaching in the classroom. The poster will highlight effective metadata standards and elements, cross-domain metadata uses, metadata mapping and implementation.

Understanding the Nature of Metadata: Systematic Review

“…This affected the complexity of metadata [ 21 ] and data quality [ 36 ], which led to the underutilization of metadata [ 36 ]. The absence of standards and thus, their nonuse created several problems: metadata were heterogeneous in structure and format and contained bad or missing descriptions, preventing the understanding of existing metadata and resulting in low quality [ 43 , 44 ]. Using different units or precision for quantitative measurements complicated the usage [ 27 ], and the heterogeneous formats prevented machine readability, which therefore worsened the identification [ 45 , 46 ], accessibility [ 47 ], retrieval [ 31 ], and validation [ 26 ].…”

“…A general problem related to every standardized data capture was the free-text elements [ 56 ]. Metadata elements also contained descriptions and definitions to understand the purpose of the items, but these included synonyms and spelling variations or naming conflicts [ 44 ], causing a data discrepancy problem if such data were shared. A viable solution was adding semantic codes to the corresponding data elements, which represented a deeper semantic understanding.…”

“…A viable solution was adding semantic codes to the corresponding data elements, which represented a deeper semantic understanding. Eichenlaub et al [ 44 ] assumed that de facto standard thesauri from research fields—in the authors’ case fashion—did not cover (commonly) used terms, or the use of proprietary codes cause semantic heterogeneity [ 17 ]. The reviewed papers proposed a better annotation process, which a domain expert or natural language processing tools [ 23 ] should execute, supplemented by postcoordination and an expert review to ensure consistent encoding [ 57 ].…”

“…Thus, metadata models or the corresponding software [ 63 ] were too complicated for health care professionals without certain necessary information technology skills [ 32 ] and therefore rarely used. In addition to the technical issues, the problem extended to the conceptual level: the model would not be clearly comprehensible if the stakeholder, users, and organizations slightly deviated in their understanding of the use cases [ 44 ]. As Varghese et al [ 55 ] aptly noted, simple disagreements about modeling decisions led to inadequate models.…”

“…As Varghese et al [ 55 ] aptly noted, simple disagreements about modeling decisions led to inadequate models. A tight feedback loop was recommended between users and the metadata curator to match the expected outcomes and a shared understanding of the metadata elements [ 44 , 64 ]. For example, extending metadata vocabularies with natural definitions would help to support the end users [ 64 ].…”