Dataset Geography: Mapping Language Data to Language Users
This addresses data bias issues in NLP for improving language diversity and coverage, though it is incremental as it focuses on analysis rather than new methods.
The study quantified the geographical representativeness of NLP datasets relative to language speakers' needs, revealing mismatches and providing insights into cross-lingual consistency and evaluation robustness.
As language technologies become more ubiquitous, there are increasing efforts towards expanding the language diversity and coverage of natural language processing (NLP) systems. Arguably, the most important factor influencing the quality of modern NLP systems is data availability. In this work, we study the geographical representativeness of NLP datasets, aiming to quantify if and by how much do NLP datasets match the expected needs of the language speakers. In doing so, we use entity recognition and linking systems, also making important observations about their cross-lingual consistency and giving suggestions for more robust evaluation. Last, we explore some geographical and economic factors that may explain the observed dataset distributions. Code and data are available here: https://github.com/ffaisal93/dataset_geography. Additional visualizations are available here: https://nlp.cs.gmu.edu/project/datasetmaps/.