Curated datasets, including synthetic and domain-specific diachronic text corpora.
Scientific abstracts (~871k) from 875 journals, from E-Research and PubMed databases.
Synthetic datasets to evaluate key dimensions of LSC, generated using LLMs and WordNet.
Articles (10.9M) from “The New York Times Developer Network” for every month in each year.
Scripts to scrape, process, and evaluate conceptual change in text corpora.
Scrape articles from “The New York Times Developer Network” by month and year, adhering to legal restrictions. (1851-yesterday).
Get SciMago-indexed psychology journals to filter databases by psychology domain.
Evaluate dimensions of LSC concurrently, and sociocultural drivers, to study conceptual change.
Evaluate concept semantic severity (weighted average linking affective ratings to target collocates).