Architectures of Meaning, A Systematic Corpus Analysis of NLP Systems
Computation and Language
2021-07-20 v1 Artificial Intelligence
Abstract
This paper proposes a novel statistical corpus analysis framework targeted towards the interpretation of Natural Language Processing (NLP) architectural patterns at scale. The proposed approach combines saturation-based lexicon construction, statistical corpus analysis methods and graph collocations to induce a synthesis representation of NLP architectural patterns from corpora. The framework is validated in the full corpus of Semeval tasks and demonstrated coherent architectural patterns which can be used to answer architectural questions on a data-driven fashion, providing a systematic mechanism to interpret a largely dynamic and exponentially growing field.
Keywords
Cite
@article{arxiv.2107.08124,
title = {Architectures of Meaning, A Systematic Corpus Analysis of NLP Systems},
author = {Oskar Wysocki and Malina Florea and Donal Landers and Andre Freitas},
journal= {arXiv preprint arXiv:2107.08124},
year = {2021}
}
Comments
20 pages, 6 figures, 9 supplementary figures, Lexicon.txt in the appendix