A Reproducibility Study on Quantifying Language Similarity: The Impact of Missing Values in the URIEL Knowledge Base
This is an incremental study that identifies reproducibility issues in a widely used tool for multilingual NLP research, affecting researchers relying on URIEL for language characterization.
The study tackled the problem of quantifying language similarity using the URIEL knowledge base, revealing that URIEL has ambiguity in calculating language distances and handling missing values, with 31% of languages lacking typological feature information, which undermines reliability, especially for low-resource languages.
In the pursuit of supporting more languages around the world, tools that characterize properties of languages play a key role in expanding the existing multilingual NLP research. In this study, we focus on a widely used typological knowledge base, URIEL, which aggregates linguistic information into numeric vectors. Specifically, we delve into the soundness and reproducibility of the approach taken by URIEL in quantifying language similarity. Our analysis reveals URIEL's ambiguity in calculating language distances and in handling missing values. Moreover, we find that URIEL does not provide any information about typological features for 31\% of the languages it represents, undermining the reliabilility of the database, particularly on low-resource languages. Our literature review suggests URIEL and lang2vec are used in papers on diverse NLP tasks, which motivates us to rigorously verify the database as the effectiveness of these works depends on the reliability of the information the tool provides.