Prototypical rhyme contains the section of the verse from the last stressed vowel on. Other, more complex, types of rhyme include more components. For the purposes of this project, the following internal structure of rhyme analysis was adopted, on which the algorithms of rhyme detection and identification were based.

First, the rhyme is categorized on the basis of the stressed vowel per verse (Line Rhyme). When this vowel is in the ultima, then it is an oxytone or masculine rhyme (masculine: M). Stressed vowels in the penult or antepenult make for feminine rhymes, which are further classified into paroxytone (feminine 2s: F2) and proparoxytone (feminine 3s: F3) rhymes.

Once initial classification is completed, further information on other aspects of the rhyme are provided, if applicable, relevant to the following parameters:

(I). Rich rhyme

(II). Imperfect rhyme

(III). Pre-rhyme identical vowel

(IV). Mosaic rhyme

(V). Copy

Notably, the type and structure of the poems is not considered – for example, the different properties of the sonnets (as in the poetry of Varnalis and Mavilis) or the rigid structure in quatrains with verse alternations of 8-7-8-7 syllables in Solomos’ “Hymn to Liberty.” These properties could further constrain what is considered a rhyme in each poet and poem. Given that the algorithm identifies rhymes in a broader sense, it thus, sometimes, finds more than one pair of rhymes for the same verse.

The ‘rhyme tables’ section, next, briefly describes parameters (I) – (V) with examples, mentions further sub-categories, and includes the abbreviations used for the corresponding rhyming patterns. For these descriptions, we follow and generally adapt definitions as given in e.g., Holtman (1996) (for Greek bibliography see: Kokolis 1993, and Stavrou 2010).