2,961 to 2,969 of 2,969 Results
Plain Text - 2.7 KB -
MD5: a2f97a0da50b51269b93b9dc00127a30
Licence and copyright notices for the bundled JavaScript library, and the statement that CC BY 4.0 covers the data tables, figures and documentation authored by the depositor. Plain text, UTF-8. |
Comma Separated Values - 11.4 KB -
MD5: 784c5b60522201398845aa3b97dea861
One row per treaty (141 rows): family assignment from average-linkage clustering on cosine distance cut at 0.65 (family_id, family_label, family_size), the signature year, and the Dataset 1 FPS typology code. Thirty-eight families, of which 24 are singletons. UTF-8, comma-separated. |
Comma Separated Values - 732.5 KB -
MD5: a2de52db4284dbaeba4f155ac34784e4
All 9,870 unordered treaty pairs with cosine similarity, Jaccard lexical overlap, the number of shared new-model clause-feature packages (0 to 9, computed from Dataset 2) and the signature-year gap. Cosine similarity correlates with the shared-package count at Pearson r = 0.55. UTF-8, comma-separated. |
Javascript Code - 628.7 KB -
MD5: 7ee044593d3ab61bec5dbfddd308cc2b
Bundled open-source JavaScript library (vis-network 10.1.0, dual licensed under MIT and Apache 2.0) included so that network_visualization.html works offline. Not a data file; keep it in the same folder as the HTML file. |
PNG Image - 59.9 KB -
MD5: c41253bfb54f01adaafcb5702cd4fd8c
Figure A: distribution of pairwise cosine similarity across all 9,870 treaty pairs. PNG. |
PNG Image - 133.7 KB -
MD5: 6918c0654f9623ab5ccd44a781281be0
Figure B: the 141 by 141 similarity matrix ordered by family assignment, in which the two dominant template blocks are visible. PNG. |
PNG Image - 169.1 KB -
MD5: a96c4121f85525da621d9d4eb7af8854
Figure C: treaty families by size and dominant model (New Turkish, Liberalization, EU-harmonization, Early, singletons). PNG. |
PNG Image - 93.9 KB -
MD5: 00c60e9cfcf1247b5115fd45979a974d
Figure D: parent versus child signature years for the 88 descent edges, showing that drafting diffuses forward in short steps. PNG. |
PNG Image - 80.6 KB -
MD5: 0319202ad674bc120f54c5f4e378d990
Figure E: textual similarity against the number of shared clause-feature packages coded in Dataset 2 (Pearson r = 0.55 across all 9,870 pairs). PNG. |
