morphemes reference

Morphemes Reference (.NET Port)

This page ports Java wiki Morphemes content to ZemberekDotNet terminology.

It is a practical reference for:

• POS and secondary POS tags
• common morpheme IDs used in analyses
• output nuances in .NET analysis formatting

Related implementation:

• TurkishMorphotactics
• SingleAnalysis

1. Primary POS IDs

Common primary POS IDs used in morphology output:

• Noun
• Adj
• Adv
• Conj
• Interj
• Verb
• Pron
• Num
• Det
• Postp
• Ques
• Dup
• Punc
• Unk (unknown, Zemberek-specific)

2. Secondary POS IDs

Common secondary POS IDs:

• Demons (demonstrative)
• Time
• Quant
• Ques
• Prop (proper noun)
• Pers (personal pronoun)
• Reflex
• Ord, Card, Percent, Ratio, Dist

Additional runtime/token categories may appear in dictionary metadata in this repository (for example date/email/url-like categories).

3. Agreement and Possessive Morphemes

Agreement:

• A1sg, A2sg, A3sg
• A1pl, A2pl, A3pl

Possessive:

• P1sg, P2sg, P3sg
• P1pl, P2pl, P3pl
• Pnon (no possession)

4. Case Morphemes

• Nom (nominative)
• Dat (dative)
• Acc (accusative)
• Abl (ablative)
• Loc (locative)
• Ins (instrumental)
• Gen (genitive)
• Equ (equative)

5. Common Derivational Morphemes

A non-exhaustive set used by current morphotactics:

• Dim, Ness, With, Without, Related, JustLike
• Agt, Become, Acquire
• Caus, Recip, Reflex, Able, Pass
• Inf1, Inf2, Inf3, ActOf
• PastPart, NarrPart, FutPart, PresPart, AorPart
• NotState, FeelLike, EverSince, Repeat, Almost, Hastily, Stay, Start
• AsIf, While, When, SinceDoingSo, AsLongAs, ByDoingSo, Adamantly
• AfterDoingSo, WithoutHavingDoneSo, WithoutBeingAbleToHaveDoneSo
• Zero

6. Verb/Tense/Polarity Morphemes

• Polarity/modality: Neg, Unable, Cop
• Tense/aspect: Pres, Past, Narr, Cond, Prog1, Prog2, Aor, Fut
• Mood/person-related: Imp, Opt, Desr, Neces

7. Output Nuance in .NET

In SingleAnalysis creation, Nom and Pnon are intentionally skipped in many formatted outputs to reduce visual noise.

See logic in:

• SingleAnalysis

This means internal morpheme state and displayed analysis text can differ in verbosity.

8. How to Inspect Morphemes in Practice

1. Analyze a word/sentence with morphology module.
2. Print lexical output (FormatLexical) for stable morpheme IDs.
3. Inspect GetMorphemes(), GetLemmas(), and GetStems() for programmatic usage.

9. Notes on Cross-System Comparison

Java/Oflazer IDs and Zemberek IDs overlap significantly, but outputs are not guaranteed to match 1:1.

When evaluating parity:

1. Compare accepted analyses, not just one string format.
2. Validate with corpus/context and disambiguation behavior.
3. Prefer repository tests as source of truth for current .NET behavior.