Metraipolis › archive › dc_6ca229da75de

Stop Removing Stopwords: How an Inherited Preprocessing Default Distorts Legal Text-as-Data — arXiv:2609.19153

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.19153
License (as archived): CC0-1.0
Archived: 2026-10-10T03:47:56.412Z
SHA-256 of the archived text: 70de86d19aa79727ee90d7306087cc5b720f642774a85ba615e267455c94c01b
Stable URL of this page: https://metraipolis.com/documento/dc_6ca229da75de

How to cite

Stop Removing Stopwords: How an Inherited Preprocessing Default Distorts Legal Text-as-Data — arXiv:2609.19153 — arXiv (en), https://arxiv.org/abs/2609.19153 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6ca229da75de (Metraipolis archive dc_6ca229da75de, sha256 70de86d19aa79727ee90d7306087cc5b720f642774a85ba615e267455c94c01b)

BibTeX

@misc{dc6ca229da75de,
  title = {Stop Removing Stopwords: How an Inherited Preprocessing Default Distorts Legal Text-as-Data — arXiv:2609.19153},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.19153},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6ca229da75de; sha256 70de86d19aa79727ee90d7306087cc5b720f642774a85ba615e267455c94c01b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.19153. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Empirical legal scholarship increasingly treats judicial text as data, and much of it still runs on sparse, interpretable pipelines (TF-IDF features and linear classifiers) because the textual feature is often the object of study rather than a means to a prediction. Yet these pipelines inherit preprocessing defaults from mid-century information retrieval that were never validated against classification accuracy. The most entrenched of these is stopword removal. This study introduces an exhaustive single-word ablation that measures a preprocessing step's effect directly against the downstream objective, and applies it to stopword removal. Matching Supreme Court Database labels to Caselaw Access Project opinion texts, the study examines two binary tasks, ideological direction (no-removal baseline F1 about 0.68) and constitutional versus non-constitutional law type (about 0.92), across 7,668 and 7,001 opinions. For each task, the ablation removes each of roughly 18,500 candidate words in turn, and a task-specific stoplist is built from the resulting measurements. Generic stoplists in common use fall below the no-removal baseline on held-out opinions in all twelve tests. The task-specific stoplists move held-out F1 by +0.0023 (95% CI [-0.0124, +0.0170]) on ideology and by +0.0001 ([-0.0082, +0.0085]) on law type. Neither task shows a detectable benefit from removal, and a supplemental analysis finds that word-level statistics predict a word's removal effect poorly, because the words' true removal effects differ by less than the measurement can register. The method generalizes to any inherited preprocessing default, and the result is a caution specific to interpretable legal text-as-data, where a step that reshapes which features a model sees can distort the doctrinal and ideological signal the research is meant to recover. The burden of proof sits with removal.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL) ; Computers and Society (cs.CY); Information Retrieval (cs.IR); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.19153 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-10T03:24:08.108Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).