Digital Trace Data Lab

Research on digital trace data uses computational methods and data science approaches to study communication phenomena like social networks, the circulation of linguistic patterns, or online behaviors

The digital trace data Lab launched in January of 2023 as a home for computational social science research that involves the use of digital trace data: information — and ways of framing information — that appears to be peripheral but can have a real impact on how people communicate, make decisions, and understand the world around them.

Housed in NDSU’s Department of Communication, research in the digital trace data lab studies communication contexts where complex information — around issues like climate change, health and medicine, or financial disclosures — circulates in and across discursive systems. In our research, we use tools and methods from deep learning-based language modeling, computational rhetoric, and traditional NLP. Our goals are to facilitate and promote research using digital trace data; bring together faculty, advanced graduate students, and other interested researchers to collaborate on such projects; and serve as a resource for NDSU researches who want to integrate digital trace data and data science methods in their own research programs.

Website: Digital Trace Data Lab

Featured Publications:
  • Majdik, Z. P., Graham, S. S., Shiva Edward, J. C., Rodriguez, S. N., Karnes, M. S., Jensen, J. T., Barbour, J. B., & Rousseau, J. F. (2024). Sample size considerations for fine-tuning large language models for named entity recognition tasks: Methodological study. JMIR AI, 3, e52095. https://doi.org/10.2196/52095

  • Majdik, Z. P., & Graham, S. S. (2024). Rhetoric of/with AI: An introduction. Rhetoric Society Quarterly, 54(3), 222–231. https://doi.org/10.1080/02773945.2024.2343264

  • Graham, S. S., Harrison, K. R., Edward, J. C. S., Majdik, Z. P., Barbour, J. B., & Rousseau, J. F. (2024). Beyond bias: Aggregate approaches to conflicts of interest research and policy in biomedical research. World Medical & Health Policy. https://doi.org/10.1002/wmh3.608

  • Majdik, Z. P., & Wynn, J. (2023). Building better machine learning models for rhetorical analyses: The use of rhetorical feature sets for training artificial neural network models. Technical Communication Quarterly 32(1), 63–78. https://doi.org/10.1080/10572252.2022.2077452

  • Lu, S. (2022). News technology innovation as a field: A structural topic modeling analysis of patient data in mainland China. Communication & Society, 59, 147-175. http://cschinese.com/issueArticle.asp?P_No=90&CA_ID=688

  • Luqiu, L. R., & Lu, S. (2021). Bounded or boundless: A case study of foreign correspondents’ use of Twitter during the 2019 Hong Kong protests. Social Media + Society, 7(1). https://doi.org/10.1177/2056305121990637

  • Masullo, G. M., Tenenboim, O., & Lu, S. (2021). “Toxic atmosphere effect”: Uncivil online comments cue negative audience perceptions of news outlet credibility. Journalism. https://doi.org/10.1177/14648849211064001.

  • Graham, S. S., Majdik, Z. P., & Clark, D. (2020). Methods for extracting relational data from unstructured texts prior to network visualization in humanities research. Journal of Open Humanities Data, 6(8). https://doi.org/10.5334/johd.21

  • Majdik, Z. P. (2019). A computational approach to assessing rhetorical effectiveness: Agentic framing of climate change in the Congressional Record, 1994–2016. Technical Communication Quarterly, 28(3), 207–222. https://doi.org/10.1080/10572252.2019.1601774

  • Lu, S., Chen, W., Li, X., & Zheng, P. (2018). The Chinese smog crisis as media event: Examining Twitter discussion of the documentary Under the Dome. Policy & Internet, 10(4), 483-508.

Top of page