AI & Visual Communication

Topic modeling of video and image data: A visual semantic unsupervised approach

VisTopics, a computational framework for analyzing large-scale visual datasets through frame extraction, deduplication, image captioning, and topic modeling.

Lokmanoglu, A. D., & Walter, D. (2025). Topic modeling of video and image data: A visual semantic unsupervised approach. Communication Methods and Measures. [DOI] [Code]

Highlights:

Lokmanoglu, A. D., & Walter, D. (2025). Topic modeling of video and image data: A visual semantic unsupervised approach. Communication Methods and Measures.

Visual Framing and AI Coding

Damanhoury, K. E., Lokmanoglu, A., Massignan, V., & Saleh, F. (2026). International media coverage of the 2023 Gaza War: A hybrid methodological approach to cross-cultural visual analysis. Media, War & Conflict. [DOI] [Code]

Damanhoury, K. E., Winkler, C., Lokmanoglu, A., & Glanz, K. A. C. (2026). Visual Framing in the AI Era: Lessons from Manual Approaches for Computational Methods. Computational Communication Research, 8(1), 1. [DOI] [Code]

Highlights:

Introduces a hybrid workflow that combines manual visual framing analysis with AI-assisted image annotation.
Demonstrates how large language models and computer vision tools can support—not replace—human coding.
Provides open-source replication code for cross-cultural visual framing and AI-assisted coding workflows.

Meme Content Analysis: Classification and Diffusion of Memes in Fringe Social

This project focuses on the classification and diffusion of memes within fringe social platforms, analyzing the content and spread of memes related to extremism and radicalization.

Highlights:

Meme content classification
Diffusion analysis in fringe platforms
Impact on radicalization

Project Details:

Start Date: 2022
End Date: 2023
Funding Agency: The Global Network on Extremism and Technology (GNET)

Visual clustering output from partisan platforms categorizing memes.

Other relevant projects and repositories

Yang, F., Cai, M., Mortenson, C., Fakhari, H., Lokmanoglu, A. D., Diakopoulos, N., Nisbet, E. C., & Kay, M. (2024). The Backstory to “Swaying the Public”: A Design Chronicle of Election Forecast Visualizations. IEEE Transactions on Visualization and Computer Graphics. [DOI]
Yang, F., Cai, M., Mortenson, C. M., Fakhari, H., Lokmanoglu, A. D., Hullman, J., Franconeri, S., Diakopoulos, N., Nisbet, E., & Kay, M. (2023). Swaying the Public? Impacts of Election Forecast Visualizations on Emotion, Trust, and Intention in the 2022 U.S. Midterms. IEEE Transactions on Visualization and Computer Graphics. (*IEEE VIS23 Best Paper Award*) [Preprint]
vistopics (Visual Semantic Topic Modeling) – Python package for analyzing video and image data [PyPI] VisTopics Project Page [Website]
Replication Code [GitHub]
Arabic and English OCR [GitHub]