Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

n-destin/Kinyarwanda-Morphology-Disambiguator

Domain:

natural language processing

Record type:

software
Creator:
n-d
Host:
Morphology generator systems require morphology disambiguation. I take a neural network approach in designing a morphology disambiguation process for Kinyarwanda (My native language). Low-resource languages benefit from word-level information (Such as morphemes) to enhance information in their embeddings. Often, the morphology generation systems require morphology disambiguation (Determining) which segmentations generated are the correct ones. In this code, I take a Neural Network approach to addressing this problem for Kinyarwanda (My native language). This code is based on the paper: Kinyarwanda Morphology disambiguation (KinyaBERT: a Morphology-awa…) ## Improvements made: - Included neighborhood information in feature extraction - Embeddings were produced using a physics-inspired superposition of the morphemes embeddings (Tensorized embeddings)

Visit

github.com

Languages

Kinyarwanda

Tags

bertlanguage-modelmachine-learningtransformer

Similar

Resolving Hiatus in Kinyarwanda MorphologyKinyaBERT: a Morphology-aware Kinyarwanda Language ModelKinyaBERT: a Morphology-aware Kinyarwanda Language ModelTongue-Height Harmony in Kinyarwanda Verbal MorphologyLes Beta Israel, un destin juif, dossier Fascinante Éthiopie juive et chrétienne, N° 235, p. 52-55Templatic morphology through syntactic selection: Valency-changing extensions in Kinyarwanda

Resolving Hiatus in Kinyarwanda Morphology

KinyaBERT: a Morphology-aware Kinyarwanda Language Model

Pre-trained language models such as BERT have been successful at tackling many natural language processing tasks. However, the unsupervised sub-word tokenization methods commonly used in these models (e.g., byte-pair encoding - BPE) are sub-optimal at handling morp

KinyaBERT: a Morphology-aware Kinyarwanda Language Model

Pre-trained language models such as BERT have been successful at tackling many natural language processing tasks. However, the unsupervised sub-word tokenization methods commonly used in these models (e.g., byte-pair encoding - BPE) are sub-optimal at handling morp

Tongue-Height Harmony in Kinyarwanda Verbal Morphology

Les Beta Israel, un destin juif, dossier Fascinante Éthiopie juive et chrétienne, N° 235, p. 52-55

International audience

Templatic morphology through syntactic selection: Valency-changing extensions in Kinyarwanda

The existence of both morphological templates (Hyman 2003) and Mirror Principle (Baker 1985) complia