AI and social science: Automatic classification tools for big data analysis in sociological research
Andrea Nucita, Assunta Penna, Antonia Cava, Giancarlo Iannizzotto, Massimo Mucciardi · 2026
Read AI and social science: Automatic classification tools for big data analysis in sociological research by Andrea Nucita, Assunta Penna, Antonia Cava, Giancarlo Iannizzotto, Massimo Mucciardi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0351336.txt.
Description
Andrea Nucita and colleagues staged a coding contest between human experts and artificial intelligence on thousands of Facebook posts from two Italian universities. Sociologists increasingly face data too big to code by hand, but categories like the purpose of an institutional post are slippery. Three researchers coded the posts independently, creating the human benchmark against which machine classifications were judged.
Ambiguity plagued several categories, and humans disagreed with each other plenty. Against that honest baseline, the automated systems agreed with human coding about as well as the humans agreed among themselves. The authors propose a hybrid bench: AI as one more coder in the team, its disagreements reviewed like anyone's. For social science facing platform-scale data, the study reframes the question from whether machines code perfectly to whether they code as reliably as we do.
Catalog record
| CREATOR | Andrea Nucita, Assunta Penna, Antonia Cava, Giancarlo Iannizzotto, Massimo Mucciardi |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43108 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |