@inproceedings{mariano-etal-2026-data,
title = "Data-Faithful Natural Language Generation for Explaining Fitness-Market Anomalies",
author = "Mariano, Kauan Divino Pouso and
Almada, Fabrycio Leite Nakano and
Monteiro, Victor Emanuel da Silva and
Dutra, Maykon Adriell and
de Faria, Jefferson Felex and
Peixoto, Jennifer Vit{\'o}ria da Silva and
Freitas, Kamylla Sejane Pouso",
editor = "Barbosa, Bryan Khelven da Silva and
Paes, Aline and
Felippo, Ariani Di",
booktitle = "Proceedings of the 17th {B}razilian Symposium in Information and Human Language Technology",
month = oct,
year = "2026",
address = "Cuiab{\'a}, Mato Grosso, Brazil",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.stil-1.50/",
doi = "10.5753/stil.2026.29472",
pages = "567--572",
abstract = "This study presents an integrated pipeline that transforms anomaly-detection outputs from global fitness-market data into data-faithful narratives. Using a 132-country panel and a 2019 baseline, we analyze observed post-COVID-19 recovery through 2025, while treating 2026 only as a modelled extension. Four complementary detectors{---}regional-median deviations, linear-regression residuals, Random Forest residuals, and Isolation Forest{---}are combined through method agreement and mapped to a rule-based typology. A deterministic, template-based Natural Language Generation (NLG) component then produces short, medium, and complete explanations from tabular evidence. In 2025, 14 countries were consensus anomalies; Guyana exemplified market growth without participatory expansion, with revenue recovery of +401.43{\%}, participation recovery of -2.06{\%}, and a 403.50-point gap. All 132 narratives passed automatic factual-consistency checks. These checks validate field preservation rather than fluency, usefulness, or explanatory quality; human evaluation remains future work."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="mariano-etal-2026-data">
<titleInfo>
<title>Data-Faithful Natural Language Generation for Explaining Fitness-Market Anomalies</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kauan</namePart>
<namePart type="given">Divino</namePart>
<namePart type="given">Pouso</namePart>
<namePart type="family">Mariano</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fabrycio</namePart>
<namePart type="given">Leite</namePart>
<namePart type="given">Nakano</namePart>
<namePart type="family">Almada</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Victor</namePart>
<namePart type="given">Emanuel</namePart>
<namePart type="given">da</namePart>
<namePart type="given">Silva</namePart>
<namePart type="family">Monteiro</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maykon</namePart>
<namePart type="given">Adriell</namePart>
<namePart type="family">Dutra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jefferson</namePart>
<namePart type="given">Felex</namePart>
<namePart type="family">de Faria</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jennifer</namePart>
<namePart type="given">Vitória</namePart>
<namePart type="given">da</namePart>
<namePart type="given">Silva</namePart>
<namePart type="family">Peixoto</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kamylla</namePart>
<namePart type="given">Sejane</namePart>
<namePart type="given">Pouso</namePart>
<namePart type="family">Freitas</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 17th Brazilian Symposium in Information and Human Language Technology</title>
</titleInfo>
<name type="personal">
<namePart type="given">Bryan</namePart>
<namePart type="given">Khelven</namePart>
<namePart type="given">da</namePart>
<namePart type="given">Silva</namePart>
<namePart type="family">Barbosa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aline</namePart>
<namePart type="family">Paes</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ariani</namePart>
<namePart type="given">Di</namePart>
<namePart type="family">Felippo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Cuiabá, Mato Grosso, Brazil</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This study presents an integrated pipeline that transforms anomaly-detection outputs from global fitness-market data into data-faithful narratives. Using a 132-country panel and a 2019 baseline, we analyze observed post-COVID-19 recovery through 2025, while treating 2026 only as a modelled extension. Four complementary detectors—regional-median deviations, linear-regression residuals, Random Forest residuals, and Isolation Forest—are combined through method agreement and mapped to a rule-based typology. A deterministic, template-based Natural Language Generation (NLG) component then produces short, medium, and complete explanations from tabular evidence. In 2025, 14 countries were consensus anomalies; Guyana exemplified market growth without participatory expansion, with revenue recovery of +401.43%, participation recovery of -2.06%, and a 403.50-point gap. All 132 narratives passed automatic factual-consistency checks. These checks validate field preservation rather than fluency, usefulness, or explanatory quality; human evaluation remains future work.</abstract>
<identifier type="citekey">mariano-etal-2026-data</identifier>
<identifier type="doi">10.5753/stil.2026.29472</identifier>
<location>
<url>https://aclanthology.org/2026.stil-1.50/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>567</start>
<end>572</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Data-Faithful Natural Language Generation for Explaining Fitness-Market Anomalies
%A Mariano, Kauan Divino Pouso
%A Almada, Fabrycio Leite Nakano
%A Monteiro, Victor Emanuel da Silva
%A Dutra, Maykon Adriell
%A de Faria, Jefferson Felex
%A Peixoto, Jennifer Vitória da Silva
%A Freitas, Kamylla Sejane Pouso
%Y Barbosa, Bryan Khelven da Silva
%Y Paes, Aline
%Y Felippo, Ariani Di
%S Proceedings of the 17th Brazilian Symposium in Information and Human Language Technology
%D 2026
%8 October
%I Association for Computational Linguistics
%C Cuiabá, Mato Grosso, Brazil
%F mariano-etal-2026-data
%X This study presents an integrated pipeline that transforms anomaly-detection outputs from global fitness-market data into data-faithful narratives. Using a 132-country panel and a 2019 baseline, we analyze observed post-COVID-19 recovery through 2025, while treating 2026 only as a modelled extension. Four complementary detectors—regional-median deviations, linear-regression residuals, Random Forest residuals, and Isolation Forest—are combined through method agreement and mapped to a rule-based typology. A deterministic, template-based Natural Language Generation (NLG) component then produces short, medium, and complete explanations from tabular evidence. In 2025, 14 countries were consensus anomalies; Guyana exemplified market growth without participatory expansion, with revenue recovery of +401.43%, participation recovery of -2.06%, and a 403.50-point gap. All 132 narratives passed automatic factual-consistency checks. These checks validate field preservation rather than fluency, usefulness, or explanatory quality; human evaluation remains future work.
%R 10.5753/stil.2026.29472
%U https://aclanthology.org/2026.stil-1.50/
%U https://doi.org/10.5753/stil.2026.29472
%P 567-572
Markdown (Informal)
[Data-Faithful Natural Language Generation for Explaining Fitness-Market Anomalies](https://aclanthology.org/2026.stil-1.50/) (Mariano et al., STIL 2026)
ACL
- Kauan Divino Pouso Mariano, Fabrycio Leite Nakano Almada, Victor Emanuel da Silva Monteiro, Maykon Adriell Dutra, Jefferson Felex de Faria, Jennifer Vitória da Silva Peixoto, and Kamylla Sejane Pouso Freitas. 2026. Data-Faithful Natural Language Generation for Explaining Fitness-Market Anomalies. In Proceedings of the 17th Brazilian Symposium in Information and Human Language Technology, pages 567–572, Cuiabá, Mato Grosso, Brazil. Association for Computational Linguistics.