NM2226 Introduction to Visual Cultural Analytics
A mixed-methods undergraduate course on studying and researching visual culture in contemporary media environments, combining qualitative interpretation, content analysis, and artificial intelligence and machine learning methods including image feature extraction, classification, clustering, and the use of large language models for visual annotation.
Course Overview
From Instagram and TikTok to the use of visuals in news stories, the contemporary media environment is increasingly visual. This course provides systematic training in studying and researching visual culture in contemporary media environments. Combining qualitative interpretation with artificial intelligence and machine learning methods, the course covers visual semiotics, discourse analysis, and content analysis alongside computational image processing, feature extraction using convolutional neural networks (CNNs), image classification and clustering, and the use of large language models (LLMs) for visual annotation. Students will critically assess the capabilities and limitations of these AI techniques. By the end of the semester, students will be able to develop a mixed-method project applying these approaches to visual data and present findings in both academic and applied contexts.
Learning Objectives
By the end of this course, students will be able to:
- Apply qualitative methods (semiotic, discourse, ethnographic) to interpret and analyze diverse visual materials, including images, films, videos, and digital media
- Implement basic quantitative and AI-based computational techniques — including human coding, content analysis, machine learning-based feature extraction using convolutional neural networks, and image classification and clustering — for analyzing visual datasets
- Design and conduct a small-scale mixed-method visual project that integrates cultural interpretation with empirical analysis
- Evaluate and demonstrate ethical awareness and reflexivity in collecting, analyzing, and presenting visual data, including critical assessment of the capabilities, limitations, and biases of AI tools used in visual research
- Communicate visual research findings clearly and effectively in written, oral, and visual formats
Prerequisites
No formal prerequisite. Advisory: AR2225 Reading Visual Images (or VCU2101 Understanding Visual Culture) is recommended before this methods-heavy course.
Level
Undergraduate
Institution
Department of Communications and New Media, National University of Singapore
Workload
| Activity | Hours/week |
|---|---|
| Lecture | 2 hrs — core concepts, method overviews, demonstrations |
| Tutorial | 1 hr — code-along labs, peer feedback, case discussions |
| Group project work | 4 hrs — data collection, analysis, write-up |
| Preparatory work | 3 hrs — readings, reflection notes, dataset familiarisation |
| Total | 10 hrs |
Required Textbooks
- Rose, G. (2022). Visual Methodologies: An Introduction to Working with Visual Materials (5th ed.). Sage.
- Manovich, L. (2020). Cultural Analytics. MIT Press. — free online
- Szeliski, R. (2022). Computer Vision: Algorithms and Applications (2nd ed.). Springer. — free online
Key Tools
| Tool | Purpose | Link |
|---|---|---|
| Python + Jupyter | Computational analysis | jupyter.org |
| OpenCV | Image processing | opencv.org |
| scikit-learn | Classification & clustering | scikit-learn.org |
| ImagePlot | Visualising image collections | lab.softwarestudies.com |
| Google Colab | Cloud computing (no setup) | colab.research.google.com |
| data.gov.sg | NLS Digitized Images dataset | data.gov.sg |
Group Project
The group project focuses on mapping cultural representations and diachronic changes in Singapore’s visual heritage, using the National Library Singapore Digitized Images dataset (via data.gov.sg). Students adopt a mixed-methods framework integrating qualitative interpretation with computational analysis to trace how visual cultures evolve across historical and institutional contexts.
- Weeks 1–2 — Formulate a cultural research question about visual change in Singapore
- Weeks 3–5 — Close visual and contextual readings of selected images
- Weeks 6–9 — Quantitative and computational analysis of visual metadata and temporal patterns
- Weeks 10–12 — Apply AI/LLMs for coding and annotation, with critical reflection on methodology and ethics
Assessment
| Component | Weight | Notes |
|---|---|---|
| Class participation | 15% | Contributions to lectures and tutorials |
| Individual lightning presentation | 15% | 5-min critical review of one visual analysis case |
| Reflective essay | 20% | 1,500 words, due Week 13 |
| Group project | 40% | Midterm presentation (10%, Week 6) + Final presentation (10%, Week 13) + Written report (20%, Week 13) |
| Peer assessment | 10% | Peer evaluation of group project contributions |
| Total | 100% | No final exam |
Schedule
| Week | Date | Topic | Materials |
|---|---|---|---|
| 1 | Week 1 | Introduction to Visual Methodologies and Cultural Analytics What does it mean to study images rigorously? Overview of qualitative, quantitative, and computational approaches to visual culture. | |
| 2 | Week 2 | Image as Cultural Data Ethics of visual research; making images as research data; cultural sampling strategies. | |
| 3 | Week 3 | Compositional Interpretation & Semiology The ‘good eye’ and formal analysis; Saussure, Barthes, and the grammar of visual signs. | |
| 4 | Week 4 | Discourse Analysis Foucauldian discourse analysis applied to visual materials; power, knowledge, and the visual. | |
| 5 | Week 5 | Audience Studies Spectators, viewers, and publics; reception theory; going beyond the image-audience relationship. | |
| 6 | Week 6 | Content Analysis Systematic coding of visual materials; intercoder reliability; codebook design. Midterm project presentations. | |
| 7 | Week 7 | Digital Methods & AI Aesthetics Scraping and archiving visual data; algorithmic images; what Instagram tells us about contemporary visual culture. | |
| 8 | Week 8 | Studying Culture at Scale From new media to more media; the science of culture; culture industry and media analytics. | |
| 9 | Week 9 | Representing Culture as Data Metadata, visual features, and cultural categories; moving from pixels to meaning. | |
| 10 | Week 10 | Image Processing & Feature Extraction Exploratory media analysis; colour histograms, brightness, composition; CNNs as feature extractors. | |
| 11 | Week 11 | Image Classification & Clustering Supervised and unsupervised approaches to visual datasets; k-means clustering; visualising collections. | |
| 12 | Week 12 | Data Visualization for Visual Research Using images to disseminate findings; image montages, timelines, and interactive displays; LLMs for coding and annotation. | |
| 13 | Week 13 | Group Project Presentations Final mixed-method project presentations. Reflective essay due. |