A Statistical Explanation of the Distribution of Sortal Classifiers in Languages of the World via Computational Classifiers

Abstract

Previous studies demonstrate that morphosyntactic plural markers and the structure of numeral systems have individually strong predictive power with regard to the usage of sortal classifiers in languages. We use these two factors as explanatory variables to train the computational classifier of random forests and evaluate the accuracy of their predictive power when selecting the existence/absence of sortal classifiers as response variable. Our results show that these two factors result in an excellent discrimination performance of random forests, even when taking into account sortal classifiers as an areal feature. However, the correlation between morphosyntactic plural markers and multiplicative bases is weaker than the correlation between sortal classifiers and plural markers plus multiplicative bases. We are thus able to provide novel insights with regard to probabilistic universals on sortal classifiers, and suggest an innovative cross-disciplinary approach to test the effect of implicational universals with computational methods.

A Methodological Review of Machine Learning in Applied Linguistics

“…Here are some linguistic studies by random forest algorithm. Her and Tang (2020), for example, ranked the feature importance by random forest to understand the predictive power of input variables. There are also some other similar cases, such as Deshors (2020b), andWiechmann andKerz (2014).…”

Tone and genes: New cross-linguistic data and methods support the weak negative effect of the “derived” allele of ASPM on tone, but not of Microcephalin

“…For (i), both experimental studies [ 25 , 26 ] fail to find any evidence for an effect of Microcephalin and, while this could be a false negative (its effect is too weak to be detected even with more than 400 participants) or simply not captured by the tasks, it is worthwhile taking it at face value and assuming that Microcephalin could have very well been a false positive in the original [ 9 ] study. For (ii), the intervening years have seen a revolution in the methods used to ask cross-cultural questions [ 27 ], ranging from the generalisation of mixed-effects/hierarchical regression [ 28 , 29 ], to the use of Bayesian methods [ 2 , 30 ], permutation/randomisation [ 31 ], and of phylogenetics [ 32 ] and machine learning [ 33 ]. Likewise, the availability and quality of linguistic (and cultural) data has dramatically improved, with databases such as WALS Online ; [ 34 ], PHOIBLE ; [ 35 ], LAPSyD and D-PLACE [ 36 ] being easily accessed by humans and machines.…”