NEMO: Frequentist Inference Approach to Constrained Linguistic Typology Feature Prediction in SIGTYP 2020 Shared Task

TL;DR

This paper introduces NEMO, a frequentist inference method using ridge regression for predicting linguistic features across languages, achieving high accuracy in the SIGTYP 2020 shared task.

Contribution

It presents a novel application of frequentist inference with simple estimators for linguistic feature prediction, ranking highly in a competitive shared task.

Findings

Achieved 0.66 micro-averaged accuracy on 149 languages

Ranked second and third in the shared task

Demonstrated effectiveness of ridge regression configurations

Abstract

This paper describes the NEMO submission to SIGTYP 2020 shared task which deals with prediction of linguistic typological features for multiple languages using the data derived from World Atlas of Language Structures (WALS). We employ frequentist inference to represent correlations between typological features and use this representation to train simple multi-class estimators that predict individual features. We describe two submitted ridge regression-based configurations which ranked second and third overall in the constrained task. Our best configuration achieved the micro-averaged accuracy score of 0.66 on 149 test languages.