Noah Diewald

There are many challenges in morphological fieldwork annotation. It heavily relies on segmentation and feature labeling (which have both practical and theoretical drawbacks), it’s time-intensive, and the annotator needs to be linguistically trained and may still annotate things inconsistently.

I collaborated on developing a workflow that relies on unsupervised and active learning grounded in word-and-paradigm morphology (WP). ML has the potential to greatly accelerate the annotation process and allow a human annotator to focus on problematic cases. At the same time, the WP approach makes for an annotation system that is word-based and relational, removing the need to make decisions about feature labeling and segmentation early in the process and allowing speakers of the language of interest to participate more actively since linguistic training is not necessary.

This is a proof of concept for the first step of the workflow. In a realistic fieldwork setting, annotators can process hundreds of forms per hour.

This site contains a database of information gathered during my field research on Wao Tededo. Wao Tededo, also spelled Wao Terero, is a language in the Ecuadorian Amazon. The database is currently a work in progress and cannot be viewed by the general public. Eventually, in the interest of open science and a commitment to making the data collected freely available to the Wao community, the majority of its contents will be viewable by the general public, following review by Wao speakers.

The files in this repository are mostly more fleshed-out examples from academic talks. They still need to be completed but provide details otherwise missing in overviews. In addition, the repository includes my current work for my thesis, "Wao Terero Lexical Suffixes." These contain my most current theoretical ideas and most advanced formal model.