github.com/m-chaves
Google Scholar Profile
CV
marianach16@gmail.com or mariana.chaves.e@outlook.com
I’m a statistician and data scientist with 7 years of working experience in data modeling, visualization, extraction, and transformation. Recently, I have been researching natural language processing techniques and text models in the context of fallacy detection and classification. I also worked in explainable and interpretable AI. In particular, with prototype-based interpretable neural networks and local interpretable model-agnostic explanations (LIME).
Key words about me: Natural Language Processing Data Science Argumentation FallaciesExplainable AI Interpretable AI Statistics Machine Learning
MSc DATA SCIENCE AND ARTIFICIAL INTELLIGENCE
Université Côte d’Azur, France | 2020 - 2022Honors Graduate (Mention Très Bien)
BACHELOR IN STATISTICS
University of Costa Rica, Costa Rica | 2013 - 2017Honors Graduate
RESEARCH ENGINEER
INRIA, WIMMICS team and I3S laboratory, France | April 2023 - PresentResearch in natural language processing (NLP) focused on models and techniques that utilize argumentation and other text data derived features to identify propaganda in social media and political debate contexts.
INTERNSHIP
I3S laboratory and INRIA, France | March 2022 - August 2022Worked on prototype-based interpretable neural networks approaches for text classification. This study lies on the framework of ANR TRACTIVE, a project that regroups researchers several disciplines for the understanding of gender representation in visual media. It involved implementing various NLP techniques and text classification models. Read the full work here.
INTERNSHIP
Université Côte d’Azur and INRIA, France | April 2021 - July 2021Worked on model agnostic interpretability methods in the ambit of images. More specifically, improving resampling process for LIME. Read the full work here.
JUNIOR DATA SCIENTIST
Walmart Supply Chain Analytics USA | October 2018 - July 2020Directed and developed data analysis projects to provide agile solutions to business partners. Main projects included an anomaly detection system, statistical sampling design, and consumption layer design.
DATA ANALYST
Walmart Supply Chain Analytics USA | January 2017 - October 2018Worked in data extraction, transformation, and visualization. Main projects included statistical process control techniques and KPI development.
DATA ANALYST
BAC Credomatic Regional Compliance Management, Costa Rica | April 2016 - December 2016Developed money laundering detection models for five Central American countries.
Highest GPA among Statistics majors for three consecutive years : 2013, 2014, 2015. UNIVERSITY OF COSTA RICA
Honorable Mention: 9th place (out of ≈33 000) on National Admission Exam. UNIVERSITY OF COSTA RICA 2012-2013
Honorable Mention: 1st place (out of ≈20 000) on national Admission Exam. TECHNOLOGICAL INSTITUTE OF COSTA RICA 2012-2013
python R LaTeX SQL
Goffredo, P., Chaves, M., Villata, S., & Cabrio, E. (2023). Argument-based Detection and Classification of Fallacies in Political Debates. In H. Bouamor, J. Pino, & K. Bali (Eds.), Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (pp. 11101–11112). Association for Computational Linguistics. https://doi.org/10.18653/v1/2023.emnlp-main.684