DeRose, Steven J. 1990.
Stochastic Methods for Resolution of Grammatical Category Ambiguity in Inflected and Uninflected Languages.
Ph.D. Dissertation. Providence, RI: Brown University Department of Cognitive and Linguistic Sciences.
This appendix originally began on page 223.
The tag system used for Greek is that of the Analytical Greek New Testament (Friberg and Friberg 1981a). The tags are coded according to the table shown below, adapted from that source. Friberg and Friberg (1981b) provide some additional information about the tagging method and procedures, though their account also contains some errors, such as asserting (p. 35) that “up to 20 percent of the forms were ambiguous,” when the actual total is nearer 40%.
As already discussed, the tags often include unusual information, generally conjoined to the fundmental tag via “+,” “&,” “/,” or other symbols, representing various relations. A tag is a string of letters; the first indicates major category, and the rest indicates subcategories.
In the following charts, the upper-case letters are those used in the tags to abbreviate the category names. All slots (represented by successive columns) for the particular major category of each word (shown in the first column) are always filled; inapplicable slots (e.g., case for non-participial verbs) are filled by a dash unless no applicable slots follow. For example, an accusative neuter singular noun would be tagged
N-AN-S. The first hyphen indicates that the word is not a pronoun, and the second indicates that (consequently) the word is not marked for person.
The very observant reader may notice that the table of tag frequencies (in Appendix 8) includes one occurrence of the tag “T.” This is apparently an error in the original tagged text.