An Empirical Study of Factors Affecting Language-Independent Models

Date
2022-01-04
Authors
Liu, Xiaotong
Xu, Anbang
Akkiraju, Rama
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Scaling existing applications and solutions to multiple human languages has traditionally proven to be difficult, mainly due to the language-dependent nature of preprocessing and feature engineering techniques employed in traditional approaches. In this work, we empirically investigate the factors affecting language-independent models built with multilingual representations, including task type, language set and data resource. On two most representative Natural Language Processing tasks --- sentence classification and sequence labeling, we show that language-independent models can be comparable to or even outperforms the models trained using monolingual data, and they are generally more effective on sentence classification. We experiment language-independent models with many different languages and show that they are more suitable for typologically similar languages. We also explore the effects of different data sizes when training and testing language-independent models, and demonstrate that they are not only suitable for high-resource languages, but also very effective in low-resource languages.
Description
Keywords
Case studies of Artificial Intelligence, Business Intelligence, Analytics Technologies for Industry Platforms, case study, multilingual ai, natural language processing, transfer learning
Citation
Rights
Access Rights
Email libraryada-l@lists.hawaii.edu if you need this content in ADA-compliant format.