Nardiena A. Pratama

School of Informatics, University of Edinburgh, Edinburgh, EH8 9AB UK

PhD Student in the AI for Biomedical Innovation CDT @ University of Edinburgh

Hello, I’m Nardiena (she/her). I am a first-year PhD student in the AI Centre for Doctoral Training (CDT) in Biomedical Innovation in the School of Informatics, University of Edinburgh. I am fortunate to be advised by Professors Ajitha Rajan and Paul Brennan.

I am interested in building machine learning solutions for applications in healthcare. More specifically, my research interests lie in medical imaging and explainability. My PhD is focused on developing explainable and transparent AI solutions for glioblastoma treatment.

Prior to starting my PhD, I worked as a data scientist at Telstra and a research assistant in the D-Lab, supervised by Professor Gianluca Demartini, at the University of Queensland (UQ) in Australia. For my undergraduate studies, I completed a joint double degree program in computer science at the University of Indonesia (UI) and UQ in 2021. I completed my Honours Year at UQ in 2023 while concurrently working at Telstra 3 days/week, working on brain MRI segmentation methods, supervised by Dr. Shekhar Chandra.

In my free time, I enjoy playing piano, listening to classical music, watching musicals, and taking hip-hop dance classes.

news

May 21, 2026	Competed as a finalist in the Research Spotlight Competition at London Hopper 2026 and won the People’s Choice Award!
Sep 08, 2025	Started my PhD at Uni of Edinburgh.
Jun 24, 2025	Presented my Visual Content Perception paper at ICWSM 2025.
Mar 15, 2025	Our full paper on Visual Content Perception was accepted to ICWSM 2025.
Dec 09, 2024	Our paper on Synthetic Data Generation for Entity Extraction was accepted to Digital Forensics Doctoral Symposium (DFDS) 2025.
Jul 10, 2024	Graduated from my Computer Science Honours Year with First-Class Honours!
Feb 02, 2022	Started as a graduate data scientist at Telstra’s graduate program.

selected publications

Perception of Visual Content: Differences Between Humans and Foundation Models

Nardiena A. Pratama, Shaoyang Fan, and Gianluca Demartini

Proceedings of the International AAAI Conference on Web and Social Media, Jun 2025

Abs DOI

Human-annotated content is often used to train machine learning (ML) models. However, recently, language and multi-modal foundational models have been used to replace and scale-up human annotator’s efforts. This study explores the similarity between human-generated and ML-generated annotations of images across diverse socio-economic contexts (RQ1) and their impact on ML model performance and bias (RQ2). We aim to understand differences in perception and identify potential biases in content interpretation. Our dataset comprises images of people from various geographical regions and income levels, covering various daily activities and home environments. ML captions and human labels show highest similarity at a low-level, i.e., types of words that appear and sentence structures, but all annotations are consistent in how they perceive images across regions. ML Captions resulted in best overall region classification performance, while ML Objects and ML Captions performed best overall for income regression. ML annotations worked best for action categories, while human input was more effective for non-action categories. These findings highlight the notion that both human and machine annotations are important, and that human-generated annotations are yet to be replaceable.