Semisupervised Learning-Based Word-Sense Disambiguation Using Word Embedding for Afaan Oromoo Language

Table 1

Summary of comparison with similar systems.

No	Title	Used algorithm and techniques	Reported results	References
1	Use of part of speech tagging for Afaan	EM and K-means and one to three context window size	ML approach with EM algorithm achieved 74.85% for annotated corpus and 70.35% for unannotated one. Hybrid approach with K-means algorithm scored 79.1% for annotated corpus and 74.85% for unannotated corpus.	[16]
	Oromo word sense modeling
2	Towards the sense disambiguation of Afan Oromo words using hybrid Approach (unsupervised machine learning and rule based)	Unsupervised machine learning	The result argued that WSD yields an accuracy of 56.2% in unsupervised machine learning and 65.5% in hybrid approach	[6]
		Approach with K-means and EM clustering algorithms
3	Amharic word sense disambiguation using wordnet	WordNet dictionary to make disambiguation solved contextual words only with small data. By applying Lesk algorithm	For Amharic WordNet with morphological analyzer and Amharic WordNet without morphological analyzer, they achieved an accuracy of 57.5% and 80%, respectively	[17]
4	Word sense disambiguation for Afaan Oromo language	Supervised machine learning techniques are applied to a corpus of Afaan Oromo language, to acquire disambiguation information automatically. It also applied Naïve Bayes theorem to find the prior probability and likelihood ratio of the sense in the given context.	For annotated corpus and 79.5% for unannotated 74.67% accuracy respectively	[5]
5	Word sense disambiguation using hybrid swarm intelligence approach	Partitional clustering (EM and K-means) algorithm was employed	The achieved result was encouraging; despite it is less resource requirement. The system yielded an accuracy of 76.05% for the unsupervised approach and 89.47% for the hybrid approach, respectively.	[18]