Table 5. Top 11 largest clusters of co-cited references (size sorting).

Cluster ID	Size	Silhouette	Mean (year)*	Label (LLR)†,▴	Label (TF*IDF)††	Label (MI)†††
#0	111	0.792	2011	Neonatal seizures	Antiepileptic drugs	Anorexia
#1	83	0.770	2001	Fetal malformations	Psychomotor development	Teratogenic
#2	60	0.974	2010	Neonatal seizures	Neonatal seizures	Epilepsy monitoring
#3	49	0.994	1998	Dentate gyrus	Preterm infants	Androgen
#4	48	0.873	2005	Pharmacokinetics	Malformations	Immunoassay
#5	46	0.936	1996	Women	Teratogenesis	Cohort studies
#6	42	0.778	1999	Pregnancy outcomes	Educational need	Substitution study
#7	29	0.948	2006	Histone deacetylase	Valproic acid	Glutamate
#8	23	0.997	1998	Valpromide	Teratogenicity	Epilepsy monitoring
#10	16	0.994	1997	Teratogenesis	Questionnaire	Epilepsy monitoring
#11	16	0.955	1996	Fetal valproate syndrome	Neuro development	Epilepsy

Notes:

^*Mean (year) represents the average publication time of the literature contained in this cluster.

^†LLR (Log-likelihood ratio) is one of the clustering label word extraction algorithm.

^††TF-IDF (Term frequency–inverse document frequency) is a commonly used weighting techniques for information retrieval and data mining, this algorithm can generate cluster labels based on the title of the citing document.

^†††MI (Mutual information) is also one of the clustering label word extraction algorithm.

^▴Clusters are referred in terms of the labels selected by log-likelihood ratio test method (LLR) in this study.