Skip to main content
. 2022 Jan 27;17(1):e0260948. doi: 10.1371/journal.pone.0260948

Table 4. Description of ACF values distributions of all contained symbols for various language samples.

Language sample Group Mean Median SD Q2.5% Q97.5%
Voynich Voy. .073 .068 .044 -.009 .189
Voynich generator [20] Voy. .070 .068 .054 -.031 .193
Bengalian (Bible) B .044 .037 .047 -.035 .160
Korean (Bible; hangul) C .044 .033 .079 -.087 .235
Arabic (Bible) A .037 .033 .040 -.035 .129
Russian (fiction w/ passages in French) D .034 .008 .063 -.030 .217
Armenian (Bible) A .029 .024 .033 -.028 .116
Kannada (Bible) B .024 .018 .039 -.045 .111
Latin (Bible) A .024 .018 .031 -.016 .097
Chinese (mix traditional and new) C .023 .012 .078 -.107 .215
Hebrew (Bible) A .020 .016 .027 -.020 .072
Mokole (Bible) A .019 .015 .028 -.033 .088
Eskimo (Bible) B .016 .011 .035 -.034 .088
Rigveda (hymns in Devanagari) D .015 .014 .046 -.076 .121
Polish (Chess Manual) A .013 .008 .028 -.033 .081
Czech (fiction) A .011 .009 .025 -.028 .066
Greek (fiction) A .010 .008 .021 -.030 .051
Italian (Calabrian: Neapolitan, fiction) A .009 .008 .024 -.034 .065
English (fiction) A .007 .007 .026 -.043 .072
Italian (fiction) A .007 .006 .025 -.040 .069
Greek (Homer) A .006 .005 .021 -.032 .049
English (Encrypt: Playfare) Enc. .006 .006 .012 -.017 .035
Latin (fiction) A .004 .002 .022 -.036 .056
Latin (Virgil; 1st Century BCE) A .003 .003 .017 -.029 .038
English (Encrypt: Vigenere) Enc. .003 .003 .014 -.024 .032
English (Encrypt: Fialka) Enc. .000 .000 .012 -.021 .023
English (Encrypt: GPG) Enc. .000 .000 .011 -.022 .022

Text samples also includes Voynich Manuscript, its imitation by Timm and Schinner [20] software, texts encrypted by various cryptographic algorithms (demanding or not a source of entropy). The purpose of sample “group” is to simplify the perception of the plots, the groups are assigned manually to highlight potential clusters.