[Retracted] Fine-Tuning Word Embeddings for Hierarchical Representation of Data Using a Corpus and a Knowledge Base for Various Machine Learning Applications

Table 5

Accuracy for unsupervised hypernym directionality (BLESS) and detection (WBLESS). Different score functions are used in the detection task.

Model	BLESS	WBLESS
		CBOW	21.03	47.96	42.15	44.18	36.45
SGNS	23.61	47.18	45.44	43.65	37.47
GloVe	51.93	46.10	46.40	47.00	51.92
R-CBOW	40.77	47.06	46.58	46.28	47.54
R-SGNS	46.35	47.06	47.72	46.28	47.54
JR	34.12	47.24	44.84	45.56	47.90
HyperVec	94.02	52.4	59.95	71.04	66.49
Poincare	40.68	55.14	50.12	54.32	49.88
LEAR	96.37	55.47	70.44	70.32	59.95
HWE	97.52	55.62	59.77	62.65	59.31