Loading...
Loading...
Contributors:
Dataset URI:
10.5281/zenodo.23287680<p>Replication material for the paper <em>When Design Taxonomies Meet PatentClaims: Evidence of Construct Mismatch Between Behavioural Design Taxonomies andPatent Claim Language</em>.</p> <p>The study tests whether two published taxonomies of attention-capture designtransfer to patent claim language. Two human coders and two locally hosted languagemodels classified the same fifty first independent claims, drawn from a corpus of31,267 granted patents held by Alphabet and Meta, with Cisco as an externalreference firm. Krippendorff's alpha was 0.142 (95% CI −0.087 to 0.412) forthe psychological construct against 0.695 (0.531 to 0.830) for a technical constructcoded on the same units by the same readers — a paired difference of 0.552(0.243 to 0.838).</p> <p>This deposit contains the numbered analysis scripts, the derived corpus indexwith provenance, <strong>all five coding rounds in raw form, including the two thatwere discarded</strong>, the three protocol versions and two prompt versions withtheir change history, the model run manifests (digest, seed, prompt version, date),the curated assignee list with a decision and rationale per entity, and the manifestthat links every number in the paper to the commit and package versions thatproduced it, together with the script that checks the manuscript against it.</p> <p>Patent data are from the PatentsView bulk datasets distributed through the USPTOOpen Data Portal (load of 10 April 2026). The raw bulk files are not redistributedhere because of their size; README.md states exactly which tables to download andfrom where. USPTO data are a work of the United States government and are notsubject to copyright in the United States.</p> <p>Licensing is mixed and declared per directory in README.md: the derived data,coding rounds and documentation are CC-BY-4.0; the scripts in 02_scripts/ and04_codebook/ are MIT; 05_manuscript/elsarticle.cls is a third-party fileredistributed unmodified under the LaTeX Project Public License 1.3c.</p> <p>See README.md for the directory layout and the exact order of execution, andCHANGELOG.md for what changed between versions, with before-and-after values.</p>
Keyword(s):
attention capture; construct validity; content analysis; inter-coder reliability; Krippendorff's alpha.
MeSH Term(s):
Grants and Funding:
No Grants or Funding Data Available
Associated Publication(s):
Supplementary Data:
No Supplementary Data Available
Related Dataset(s):
No Related Dataset Data Available
The National Library of Medicine (NLM) provides information about biomedical datasets. Inclusion on an NLM Web page does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: NLM Web Policies (nih.gov)