@inproceedings{ribeiro-etal-2026-portho,
    title = "Portho: A Corpus-Based Resource of Orthographic Neighbors in {E}uropean {P}ortuguese",
    author = "Ribeiro, Eug{\'e}nio  and
      Antunes, David  and
      Mamede, Nuno  and
      Baptista, Jorge",
    editor = "Souza, Marlo  and
      de-Dios-Flores, Iria  and
      Santos, Diana  and
      Freitas, Larissa  and
      Souza, Jackson Wilke da Cruz  and
      Ribeiro, Eug{\'e}nio",
    booktitle = "Proceedings of the 17th International Conference on Computational Processing of {P}ortuguese ({PROPOR} 2026) - Vol. 1",
    month = apr,
    year = "2026",
    address = "Salvador, Brazil",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-dnd/2026.propor-1.40/",
    pages = "406--415",
    ISBN = "979-8-89176-387-6",
    abstract = "Orthographic neighbors (ONs) play a central role in models of visual word recognition and have been shown to influence reading speed, lexical access, and literacy development. Despite their importance, resources providing detailed and flexible ON information remain scarce for European Portuguese. This paper introduces Portho, a corpus-based lexical resource that provides multiple ON metrics for over 43,000 word forms, using several ON definitions. In addition to classical neighborhood size measures, Portho provides frequency-based statistics and graded orthographic distance (OD) features. We analyze the statistical properties of the resource and evaluate its empirical utility in automatic text complexity assessment using the iRead4Skills corpus. Results show that while ON features alone are insufficient to predict readability, they contribute complementary information and compare favorably with existing resources for Portuguese. Portho is made publicly available in different formats to support research in psycholinguistics, readability modeling, and Natural Language Processing (NLP) for Portuguese."
}