Alexa Frankelis Portfolio 2026

Alexa Frankelis: Data Science & Archives

ABOUT ME

:afran190@pratt.edu
:LinkedIn
:Github
:https://pratt.academia.edu/AlexaJadeFrankelis

As a candidate in the M.S. in Library and Information Science program at Pratt Institute, I look through the lens of metadata management, information design, and visualization to consider how these information systems can work to either dismantle barriers toward access, or how they fortify them, and if the latter, how we can work to breakdown these structures for more equitable access.
I am an information systems professional who shows up for my community as I learn about user groups and provide reference, description, and instruction services based on their lived experiences. Whether I am building metadata schemas, designing digital collections, or providing access to information, I work so that my systems and structures are generative places. Not just where users can find themselves reflected back, but take part in as well.

CONTACT

RESUME:

Education

M.S. Library & Information Science
December 2026 (Expected)
Pratt Institute

B.A. Art History & Criticism (Honors)
August 2020
Stony Brook University

B.F.A. Photography & Video (Dean’s List)
September 2015 - September 2018 (Degree not earned) School of Visual Arts

Employment

Senior Manager, LIFE Picture Collection, People Inc. (Formerly DotDash Meredith)
New York, NY
March 2025-Present

Manager, LIFE Picture Collection, DotDash Meredith New York, NY
March 2023-March 2025

Specialist, LIFE Picture Collection, DotDash Meredith
New York, NY
August 2021-March 2023

Art Handler, The New York Public Library
New York, NY
October 2019-July 2021

Visitor Services Assistant,
The Morgan Library and Museum New York, NY
March 2019-October 2019

Certificates

Silver Certificate, Artificial Intelligence & Digital Asset Management
April 2024
Rutgers University

Advanced Certificate, Advanced Digital Asset Management January 2023
Rutgers University

Certificate, Digital Asset Management
October 2022
Rutgers University

Selected Publications

2025 Performing Intelligence: A Review of Simone Natale’s Book Deceitful Media: Artificial: Intelligence and Social Life after the Turing Test, Issue 3 (Fall / Winter 2025), NYU Bridging Fields Journal

2023 The Guided Hand: Josefa Tolrà- Madge Gill; Visionary women, Museu Nacional d'art de Catalunya

2023 The Ghostly Other (Journal Article), Visionary
Women Art x Bric-a-Brac Magazine

2022 Sleepy Hollow: Then & Now, LIFE.com article (photography also featured)

2022 Revisiting the Sites of the Salem Witch Trials,
LIFE.com article

2021 Spirits & Suffrage, Feminine Macabre Vol. III

2021 The Ghostly Other (Essay), Issue 04, Fall 2021,
Speculative Arts Research Journal

2021 The “Ordinary” Witches of LIFE Magazine,
LIFE.com article

2021 Art and Occultism, The Vindication of The Feminist
Voice Since the End of the 19th Century, “Speaking
Martian” Collaborative Series, A*Desk Magazine x
Visionary Women Art

2020 The Gendered Narrative of Witch-Hunting Through
the Centuries, Crisis and Catharsis Journal, Stony
Brook University
(American Historical Association blog feature)

Selected Presentations / Conferences

2021 The Ghostly Other (Lecture), Association for Art
History Annual, Association for Art History Annual
Conference, University of Birmingham,
Birmingham, England

2020 Senior Zoom Salon, Paul W. Zuccaire Gallery, Stony Brook University, Stony Brook, NY

2020 Spirits & Suffrage, CFP for the Emerging
Perspectives | UG and PG Masters Conference,
University of Edinburgh, Scotland

Awards

2020 Stony Brook University, Staller Award For Excellence, Stony Brook, NY

2020 University of Edinburgh, Grant Recipient, Edinburgh, Scotland

Last Updated 09.20.2026

PORTFOLIO: The selected projects below explore how metadata schemas and controlled vocabularies impact discoveries and representation: whose stories are told through what frames. My projects examine how to negotiate between institutional expectations and community-driven knowledge organization by prioritizing transparency, inclusivity, and ethical responsibility. I consider equitable metadata a technical and moral obligation, that necessitates continual collaboration, reflexivity, and responsiveness to the needs of users. My goal through these projects, is to contribute to information spaces that serve as dynamic, equitable ecosystems of knowledge and connection.

1. Project Title: LIFE Photographer Assignment Return Management System SQL Database: Design + Implementation

(Please click on the title above to be directed to my GitHub project repository page)

Project Description: INFO 639 had students design and construct a two-tier relational database to solve a real-world information management problem. I decided to develop a framework to manage the retrieval of photographic assignment materials not owned by LIFE from an external archival facility (GRM) on behalf of the photographers and estates who do own them. The project moved from a conceptual and logical entity-relationship model to a fully implemented physical schema populated with AI-generated sample data (for company information privacy) from LIFE photographers’ legacy archives.
Methods: First, I defined the business rules for our material returns process. Then, I designed the system using Crow’s Foot notation to create conceptual and logical ERDs. Finally, I implemented the physical schema in MySQL using DBeaver. I authored and tweaked SQL to conform to the completed schema, filled the database with INSERT statements using historic LIFE photographers as sample entries, and created sixteen queries demonstrating JOINs, subqueries, views, triggers, and transactions to satisfy the database's operational needs.
My Role: Sole designer and coder (Individual Project)
Learning Outcome Achieved 1): Technology
Rationale 1): This project required me to move from a basic understanding of digital systems to a more advanced technical domain required by the Technology outcome, which is to create, standardize, and implement a relational database management system from the ground up rather than using a pre-built solution. Working through two separate design phases required me to translate a vague real-world business problem, namely the uncoordinated and unrecorded return of assignment materials from an external site, into a rigid logical framework that could enforce referential integrity and business rules via programmatic means, using triggers and transactions instead of manual oversight. I had to consider how future users, specifically archive staff working with return requests, would want to access and use this data, as well as the need to diagnose and revise my SQL when the schema did not initially support those means of retrieval. Writing sixteen queries that included JOINs, subqueries, and views required all of this careful consideration. This process of iterative design, implementation, and revision is the applied technical problem solving which the rubric highlights as important for this outcome: not merely using a database system, but showing the ability to construct and modify one for a specific archival requirement.
Learning Outcome Achieved 2): User-Centered Services
Rationale 2): Although the deliverable of this project is of a technical nature, the design was always informed by the needs of two distinct user groups: archive and licensing personnel who would use the system to process returns, and photographers and estates, many of whom have legal and personal interests in the materials that represent their creative endeavors. Before we could build a database, we had to understand the needs of each group. Staff needed a solid, verifiable way to track chain of custody and return status, while rights holders wanted a return process that was accurate, timely, and sensitive to the nature of the material, especially when it involved estates of deceased photographers. My approach was to create business rules and query logic that took into account real-world use cases rather than a generic schema. This allowed the system’s outputs and eventually the interface for staff to clearly communicate return status to non-technical rights holders. This project demonstrated my ability to translate understanding of the unique informational needs of a diverse user base into the basic structure of a system, proving that user-centric concerns are just as relevant to backend database design as they are to the more obviously visible services.

2. Project Title: Visible Death, Invisible Data: Mapping Morality and its Gaps Accross Cities

(Please click on the title above to be directed to my GitHub static website and click on the image for my final presentation)

Project Description: For my INFO 664 (Programming for Cultural Heritage) final project, I researched how causes of death for people who are imprisoned are either recorded, or excluded, across open data portals in four major U.S. cities, New York, Chicago, San Francisco, and Austin. Rather than building a single unified dataset, I designed the project to generate a comparative narrative across cities, focusing on three categories: deaths of people who are incarcerated, drug-related deaths, and COVID-19 deaths. The project's central premise was that what a city's open data infrastructure fails to record is as analytically significant as what it does record, and that these omissions are not neutral but reflect political and institutional choices about whose deaths are made visible.
Methods: I retrieved data from each city's Socrata powered open data portals’ API’s in using the Python library sodapy, authenticating with an App Token and querying each dataset by its unique identifier. I built a template Jupyter notebook, using New York City as the primary case study because it was the only city that had at least partial data in all three research areas, and then customized that template for Chicago, San Francisco, and Austin. After exported the data, I then cleaned each dataset in OpenRefine, normalizing timestamp, numerical formats, and column headings, and filtering for suitable geographic specificity. Then I used the python libraries, pandas and Plot.ly to analyze and visualize the refined data with pie, bar and line graphs for each city dataset. At the same time, I built and maintained a separate tracking spreadsheet that documented the presence of data, the inclusion of demographic analyzes, and any irregularities or deficiencies I detected in each city and category, which I then visualized in a dedicated notebook.
My Role: I was the sole author and researcher on this project. (Individual Project)
Learning Outcome Achieved: Research
Rationale: This project was structured around an original research question, what mortality data is made public and what is deliberately or structurally left out. I selected my four cities specifically because their reputations for civic transparency and progressive open data policy made them a meaningful test case for that question, rather than an easy confirmation of it. My initial hypothesis that cities lauded for their transparency would be sharing detailed mortality statistics was conclusively refuted by my findings, which I considered a major finding rather than a limitation of the study. I employed a mixed-methods approach, combining quantitative analysis of existing data sets using python’s pandas and Plot.ly libraries,and a systematic qualitative documentation of absence using my self-developed city tracker, to reach conclusions neither could individually. The two-pronged method allowed me to show, for example, that New York City’s Local Law 11 theoretically requires open data to be published, but in reality, the number of people who are imprisoned deaths’ at Rikers Island was just 22 over five years, an astoundingly low figure when compared to what the public has been told about the conditions inside the facility. Arriving at that conclusion required going beyond the data provided to consider what its own incompleteness meant for institutional accountability. I used primary sources throughout, including the open data legislation I analyzed, and capped the project by detailing specific opportunities for future research including a comparative policy analysis, oversight mechanisms, and a publicly available transparency dashboard.

3. Project Title: Performing Intelligence: A Review of Simone Natale’s Book Deceitful Media: Artificial: Intelligence and Social Life after the Turing Test (Book Review)

(Please click on the title above to be directed to my book review published in NYU’s Bridging Fields and click on the image for a PDF Verison)

Project Description: As part of INFO 601 (Foundations of Information), I wrote a scholarly book review of 2,000-3,000 words on Simone Natale's, Deceitful Media, about the ways in which artificial intelligence systems construct impressions of sentience and intelligence through design, historical context, and cultural storytelling rather than actual cognitive ability. The task was to synthesize the main contributions of the book, to assess its strengths, and to place it in context of the current relevant literature with consistent academic referencing. I concentrated my analysis on parts that discuss the historical origins of AI’s “performative intelligence,” cultural narratives that maintain myths about AI, and ethical implications of users’ complicity in these lies.
Methods: I read Deceitful Media closely and isolated the three chapters that form the core of Natale’s argument: the historical story of the Turing Test and early systems like ELIZA and PARRY, the cultural critique of anthropomorphized AI in media and everyday life, and the ethical critique of user agency in the face of AI’s deceptive behavior. To provide context and a rigorous evaluation of the book’s claims, I supplemented Natale’s thesis with relevant literature, including Kate Crawford’s Atlas of AI, research on gendered AI assistants by Chin-Rothmann and Robison, and a cross-cultural study of AI perceptions by Ge et al. I used MLA style for citation of all sources. and created my own critical analysis of the strong and weak points of the book.
My Role: I was the sole author of this review. (Individual Project)
Learning Outcome Achieved: Ethical/Creative/Critical practice
Rationale: This project pushed me to move beyond simple summary to true critical analysis, identifying not only Natale’s claims, but also the weaknesses in his argument and the implications of those for information professionals’ interactions with AI. I situated Natale's central claim, that AI's appearance of intelligence is a designed performance rather than a demonstrated capability, within its stakes for LIS practice specifically, connecting his analysis of performative deception to the information literacy, ethical metadata, and critical technological stewardship responsibilities that increasingly define the profession as AI systems mediate access to knowledge. Instead of simply agreeing with Natale’s framework, I argued that there is a big limitation to his analysis, which is very much based on Western cultural references. I applied the cross-cultural study by Ge et al., which showed that the social relevance of AI is not general, but culturally constructed. This is something that Natale’s otherwise comprehensive critique does not touch upon. I extended this analytical lens by synthesizing Natale’s analysis of anthropomorphized AI assistants with Chin-Rothmann and Robison’s findings on the perpetuation of harmful stereotypes through gendered AI design. This project showcases my ability to critically and creatively engage with contemporary technology and to situate such engagement in a professional and ethical context relevant to Library and Information Science.

4. Project Title: Metadata Application Profile for the Photo Archive of the Berenson Library at Villa I Tatti

(Please click on the title above to be directed to our CollectionBuilder website and click on the image for our final report)

Project Description: For INFO 663 (Metadata Design) my colleagues Veneranda Aguirre, Emilee Buytkins, Meg Carroll and I developed a comprehensive metadata application profile for the Photo Archive of the Berenson Library at Harvard’s Villa I Tatti. The collection has historic images of Italian Renaissance art from the collection of art historian Bernard Berenson, many of which are now missing. We built the MAP with VRA Core 4.0. We developed an entity model that separates Artwork and Photograph records, with cross-references through unique identifiers. We developed the full supporting infrastructure: functional specifications, a controlled vocabulary element set, a crosswalk to Dublin Core and Schema.org, a live Airtable metadata entry system, cataloging specifications, and a GitHub-hosted site optimized for the project's CollectionBuilder implementation on Github.
Methods: Working with my team, we established the functional specifications and tenets of the project, identified user needs (especially for art historians researching works of art or the history of photography in search, browse, view, and sort), and articulated fundamental values around accessibility, historical fidelity, transparency, and ethical representation of Berenson's archival resources. I then helped develop the element set, defining the label, data type, cardinality and required controlled vocabularies for each field (ULAN for artists, LCNAF and AAT for subjects, Wikidata for photographers) and worked on the crosswalk mapping of our VRA Core elements to their equivalents in CollectionBuilder, Schema.org and Dublin Core Terms to ensure interoperability. Emilee and I developed the Airtable-based metadata entry form that enabled the element set for catalogers, and built the GitHub site that supports our CollectionBuilder implementation, linking object locations and identifiers to their descriptive metadata records.
My Role: This was a team project completed with three other students. I contributed specifically to the functional requirements and project principles, the element set, the crosswalk, the Airtable metadata entry mechanism, and the GitHub site configuration, all developed in close partnership with my teammates. A separate team of catalogers used the guidelines and entry form we built to complete the individual catalog records.
Learning Outcome Achieved: Foundations of Library & Information Studies
Rationale: The project was an exercise in applying core Library and Information Science principles (collection, organization, and access) in an authentic archival setting under specific constraints, namely that all descriptive metadata must be derived solely from the inscriptions on the photographs, without external validation, to maintain the authenticity of Berenson’s original research. I wanted to make sure that our final structure was really supporting the way that art historians explore and navigate archival photograph collections, not imposing an arbitrary framework on the content. Since functional requirements took precedence, I had to think about how to organize information from the perspective of the user before I could make any schema decisions. The development of the element set and crosswalk demonstrated a strong creative use of basic LIS concepts: grounding each field in an appropriate controlled vocabulary, correctly distinguishing between mandatory and optional, and repeatable and non-repeatable elements, and ensuring that the resulting metadata could be used in Dublin Core, Schema.org and CollectionBuilder rather than being limited to our own system. This design effort was a viable access point that bridged the gap between information organization and its accessibility to end users through the setup of the Airtable entry system and GitHub site. This effort jointly demonstrates a deep understanding of context and environment, adjusting traditional metadata practices to the particular constraints of an art historical photo archive, while demonstrating the full arc of the Foundations outcome: the accumulation, characterization, and accessibility of an authentic archival collection according to professional standards.