Feature-Refined Unsupervised Model for Loanword Detection
This addresses the challenge of loanword detection for historical linguists by avoiding circularity from external information, though it is incremental as it builds on prior work with a hybrid approach.
The paper tackled the problem of detecting loanwords in languages by proposing an unsupervised method that uses only language-internal information, and it outperformed baseline methods on datasets from six Indo-European languages with strong performance gains in cross-linguistic scaling.
We propose an unsupervised method for detecting loanwords i.e., words borrowed from one language into another. While prior work has primarily relied on language-external information to identify loanwords, such approaches can introduce circularity and constraints into the historical linguistics workflow. In contrast, our model relies solely on language-internal information to process both native and borrowed words in monolingual and multilingual wordlists. By extracting pertinent linguistic features, scoring them, and mapping them probabilistically, we iteratively refine initial results by identifying and generalizing from emerging patterns until convergence. This hybrid approach leverages both linguistic and statistical cues to guide the discovery process. We evaluate our method on the task of isolating loanwords in datasets from six standard Indo-European languages: English, German, French, Italian, Spanish, and Portuguese. Experimental results demonstrate that our model outperforms baseline methods, with strong performance gains observed when scaling to cross-linguistic data.