Ciara Zogheib

Ciara Zogheib

Auto-generated profile, based on publicly available information

Ciara Zogheib

University of Toronto, Toronto, Ontario, Canada

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 June 2024
Verified
Conference Paper
Open access
EBEshta Bhardwaj
Eshta Bhardwaj
HGHarshit Gujral
Harshit Gujral
SWSiyi Wu
Siyi Wu
CZCiara Zogheib
Ciara Zogheib
TMTegan Maharaj
Tegan Maharaj
CBChristoph Becker
Christoph Becker
Studies of dataset development in machine learning call for greater attention to the data practices that make model development possible and shape its outcomes. Many argue that the adoption of theory and practices from archives and data curation fields can support greater fairness, accountability, transparency, and more ethical machine learning. In response, this paper examines data practices in machine learning dataset development through the lens of data curation. We evaluate data practices in machine learning as data curation practices. To do so, we develop a framework for evaluating machine learning datasets using data curation concepts and principles through a rubric. Through a mixed-methods analysis of evaluation results for 25 ML datasets, we study the feasibility of data curation principles to be adopted for machine learning data work in practice and explore how data curation is currently performed. We find that researchers in machine learning, which often emphasizes model development, struggle to apply standard data curation principles. Our findings illustrate difficulties at the intersection of these fields, such as evaluating dimensions that have shared terms in both fields but non-shared meanings, a high degree of interpretative flexibility in adapting concepts without prescriptive restrictions, obstacles in limiting the depth of data curation expertise needed to apply the rubric, and challenges in scoping the extent of documentation dataset creators are responsible for. We propose ways to address these challenges and develop an overall framework for evaluation that outlines how data curation concepts and methods can inform machine learning data practices.

27 July 2021 in Advances in Archaeological Practice
Verified
Journal Article
Open access
PDF file
DCDaniel A. Contreras
Daniel A. Contreras
ZBZachary Batist
Zachary Batist
CZCiara Zogheib
Ciara Zogheib
TCTristan Carter
Tristan Carter
The documentation and analysis of archaeological lithics must navigate a basic tension between examining and recording data on individual artifacts or on aggregates of artifacts. This poses a challenge both for artifact processing and for database construction. We present here an R Shiny solution that enables lithic analysts to enter data for both individual artifacts and aggregates of artifacts while maintaining a robust yet flexible data structure. This takes the form of a browser-based database interface that uses R to query existing data and transform new data as necessary so that users entering data of varying resolutions still produce data structured around individual artifacts. We demonstrate the function and efficacy of this tool (termed the Queryable Artifact Recording Interface [QuARI]) using the example of the Stelida Naxos Archaeological Project (SNAP), which, focused on a Paleolithic and Mesolithic chert quarry, has necessarily confronted challenges of processing and analyzing large quantities of lithic material. La documentación y análisis de materiales líticos tiene que navegar una tensión entre examinar y documentar artefactos individuales o sumas de artefactos. Esta tensión presenta un desafío tanto para el procesamiento de artefactos como para la construcción de bases de datos. Aquí presentamos una solución realizado en R Shiny lo cual hace posible que los analistas de líticos puedan introducir datos para artefactos individuales y sumas de artefactos, mientras se mantiene una estructura de datos robusta pero flexible. Esta estructura resulta ser un interfaz en el navegador que emplea R para interrogar datos existentes y transformar datos nuevos cuando sea necesario, para que los usuarios —mientras introduzcan datos de resoluciones varias— puedan producir datos con una estructura basada en el artefacto individual. Demostramos la herramienta (denominada Queryable Artifact Recording Interface [QuARI]) tras el ejemplo del Proyecto Arqueológico Stelida Naxos (SNAP, por sus signos en ingles), lo cual, por enfocarse en una cantera lítica paleolítica y mesolítica, necesariamente se ha enfrentado con los desafíos de procesar y analizar cantidades grandes de material lítico.

1 May 2025 in Journal of Eastern Mediterranean Archaeology and Heritage Studies
Verified
Journal Article
Closed access
TCTristan Carter
Tristan Carter
DCDaniel A. Contreras
Daniel A. Contreras
DMDanica D. Mihailović
Danica D. Mihailović
TMTheodora Moutsiou
Theodora Moutsiou
YPYorgan Pitt
Yorgan Pitt
CZCiara Zogheib
Ciara Zogheib
This article provides the first detailed overview of Middle Paleolithic activity at the Naxian chert source of Stelida, based on an analysis of 780 artifacts collected from the 2013–2014 survey. While several Eurasian Middle Paleolithic lithic sources have been documented, the activity at most of these sites relates almost exclusively to resource extraction and the initial stages of tool production. The Middle Paleolithic material from Stelida reflects a wider range of hominin behavior, including not only evidence for various knapping traditions (not least Levallois and discoidal core technologies) but also two concentrations of target products, including retouched tool types. The article argues that this greater breadth of practice relates to Stelida’s landscape affordances, namely the presence of springs and rock shelters that facilitated the establishment of seasonal camps, where those procuring chert likely engaged in food preparation, consumption, and tool maintenance.

14 June 2026 in Journal of the Association for Information Science and Technology
Verified
Journal Article
Open access
PDF file
CZCiara Zogheib
Ciara Zogheib
Cross‐disciplinary research is a priority for many academic institutions, with a growing body of scholarship dedicated to studying the central practice of cross‐disciplinarity: integration, or the synthesis of knowledge, information, and data across disciplines and domains. Discussions on this topic are regularly published in JASIST and ARIST and are of interest to ARIST readership because of the centrality of information to definitions of integration. Emerging challenges in this area are that (1) cross‐disciplinary integration practices are not only discussed by information scholars, but in fragmented, rarely overlapping discourses across domains, and (2) despite being linked to information, cross‐disciplinary integration is rarely explicitly positioned as an information practice. In this review, to respond to these challenges, I adopt critical interpretive synthesis methods to review the literature on cross‐disciplinary integration. I consolidate definitions of integration across 12 areas of scholarship and highlight the roles of information in these definitions. I find that certain areas of work are more likely to describe integration as the homogenization of disparate datasets, while others describe integration as a synthesis of knowledge or perspectives, and that the relationship between these integrative activities—for example, how combination of multi‐source datasets can support interdisciplinary conclusions—is rarely explored in depth. I identify concepts from the information sciences that offer ways forward to address this knowledge gap, and suggest roles for information practitioners in supporting integrative work.