Altigran Soares da Silva

Instituto de Computação
Universidade Federal do Amazonas (UFAM)
Professor

About Me

Altigran Soares da Silva is a Full Professor at the Instituto de Computação at the Universidade Federal do Amazonas (IComp/UFAM). He earned his Ph.D. in Computer Science from UFMG in 2002. His research interests include Data Management, Information Retrieval, Data Mining, Machine Learning, and Language Models, focusing on the World Wide Web, Social Media, and applications in law, finance, and health. Dr. da Silva has coordinated and participated in dozens of research projects resulting in over 150 scientific publications in journals and conference proceedings. He served as Vice Provost for Research and Graduate Studies at UFAM (2007/2009), coordinator of CA-CC at CNPq (2023/2024), and adjunct coordinator of the computing area at CAPES (2011/2013). He was also a board member (2005/2015) and council member (2016/2019) of SBC. He co-founded companies such as Akwan (acquired by Google in 2005), Neemu (acquired by Linx Systems in 2015), and Teewa (acquired by JusBrasil in 2019).

Research Interests

Data Management, Data Engineering, Data Mining
Machine Learning, Information Retrieval, Language Models
Web and Social Media

Selected Publications (2024)

Journal Papers

[Vianna@AIL24] Daniela Vianna, Edleno Silva de Moura, Altigran S. da Silva: A topic discovery approach for unsupervised organization of legal document collections. Artif. Intell. Law 32(4): 1045-1074 (2024)🎙️Podcast-PT 🎙️Podcast-ENG
Ariel Afonso, Paulo Martins, Altigran S. da Silva: SEREIA: document store exploration through keywords. Knowl. Inf. Syst. 66(10): 6101-6132 (2024)
Edleno Silva de Moura, Berg Ferreira, Altigran S. da Silva, Ricardo Baeza-Yates: BWBEV: A Bitwise Query Processing Algorithm for Approximate Prefix Search. J. Braz. Comput. Soc. 30(1): 527-541 (2024)
Conference Papers
[Silva@ReSys-Demos’24] Eduardo Alves da Silva, Leandro Balby Marinho, Edleno Silva de Moura, Altigran S. da Silva: A Tool for Explainable Pension Fund Recommendations using Large Language Models. RecSys 2024: 1184-1186 🎙️Podcast-PT 🎙️Podcast-ENG
Johny Moreira, Altigran S. da Silva, Edleno Silva de Moura, Leandro Bezerra Marinho: A Study on Unsupervised Question and Answer Generation for Legal Information Retrieval and Precedents Understanding. SIGIR 2024: 2865-2869
Gabriel Assis, Daniela Vianna, Gisele L. Pappa, Alexandre Plastino, Wagner Meira Jr, Altigran S. da Silva, Aline Paes. Analysis of Material Facts on Financial Assets: A Generative AI Approach. Proceedings of the Joint Workshop of the 7th Financial Technology and Natural Language Processing, the 5th Knowledge Discovery from Unstructured Data in Financial Services, and the 4th Workshop on Economics and Natural Language Processing @ LREC-C, 2024.
Tarsis Azevedo and Altigran Silva. 2024. Um Estudo sobre Ensino de Engenharia de Dados nas Universidades Brasileiras: Estado Atual e Perspectivas de Mercado (A Study on Data Engineering Teaching in Brazilian Universities: Current Status and Market Perspectives). In Anais do IV Simpósio Brasileiro de Educação em Computação 375-383, 2024
Marcos Lima, Eduardo Silva, and Altigran S. da Silva. 2024. Um Estudo sobre o uso de Modelos de Linguagem Abertos na Tarefa de Recomendação de Próximo Item (A Study on Open Language Models for Next-Item Recommendation). In Anais do XXXIX Simpósio Brasileiro de Bancos de Dados, 510-522, 2024
Mateus Albuquerque, Luciano Barbosa, Johny Moreira, Altigran S. da Silva, and Tiago Melo. 2024. Fine-tuning Open-source Large Language Models for Automated Response to Customer Feedback. In Anais do XII Symposium on Knowledge Discovery, Mining and Learning, 65-72, 2024
Preprints
Arthur Elwing Torres, Edleno Silva de Moura, Altigran Soares da Silva, Mario A. Nascimento, Filipe de Sá Mesquita: An Experimental Study on Data Augmentation Techniques for Named Entity Recognition on Low-Resource Domains. CoRR abs/2411.14551 (2024)
Technical Reports
Recomendações para o avanço da inteligência artificial no Brasil (Recommendations for Advancing Artificial Intelligence in Brazil) - Academia Brasileira de Ciências (ABC), 2024
Plano de Inteligência Artificial da Sociedade Brasileira de Computação (Brazilian Computer Society Artificial Intelligence Plan). Sociedade Brasileira de Computação (SBC), 2024.

Current Research Projects

Projects I am coordinating

Semantic Discovery and Explainability of Relationships in Data Lakes (Semantic Discovery and Explainability of Relationships in Data Lakes) - Focuses on creating integrated solutions for identifying and explaining semantic relationships in large data repositories, improving efficiency and interpretability. Funded by FAPEAM.
Integrating LLMs into Financial Recommendation Systems: Personalization and Bias Mitigation (Integrando LLMs em Sistemas de Recomendação Financeiros: Personalização e Mitigação de Viés) - This project aims to develop advanced methods to enhance fairness and personalization in financial recommendation systems by leveraging Large Language Models (LLMs). Funded by CNPq.
Neural Bond - Using Neural Language Models for Smart Social Media Engagement (Neural Bond - Uso de Modelos de Linguagem Neurais para Engajamento Inteligente de Usuários em Redes Sociais) - Investigates neural language models to optimize user engagement and interaction in social media platforms. Funded by FAPEAM.
Research on Methods and Techniques for Query Suggestion Systems (Pesquisa em Métodos e Técnicas para Sistemas de Sugestões de Consultas) - This project aims to study and develop new methods for relevance ranking in autocomplete systems using machine learning. The development process involves building labeled datasets and defining features to train algorithms that approximate optimal solutions. Funded by Jusbrasil (GOSHME)
Research Networks I am participating in
CIIA-Saúde - Center for Innovation in Artificial Intelligence for Health (Centro de Inovação em Inteligência Artificial para a Saúde) - Principal Investigator - A multidisciplinary project exploring AI-driven solutions to enhance diagnostics, treatment plans, and healthcare management - Funded by FAPESP/MCTI, FAPEMIG, and UNIMED-BH
IAIA - National Institute of Science and Technology in Artificial Intelligence (Instituto Nacional de Ciência e Tecnologia em Inteligência Artificial) - Steering Committee Member - National initiative aiming to advance AI technologies across multiple domains, fostering innovation and collaboration - Funded by CNPq

Current Supervisions

Ph.D. Students

Júnio da Silva de Freitas - Automatic View Generation for Databases: Enhancing SQL-to-Text Conversion with LLMs (Geração Automática de Visões para Bancos de Dados: Uma Nova Estratégia para Aprimorar a Conversão de Texto para SQL com LLMs).
Eduardo Alves da Silva - A Study on Portfolio Recommendation Methods for Pension Investments (Um Estudo sobre Métodos de Recomendação de Portfólio para Investimentos em Previdência).
Arthur Elwing Torres - Analysis of Data Augmentation in Named Entity Extraction for Domain-Specific Texts (Análise da Utilização de Aumento de Dados em Extração de Entidades Nomeadas de Textos de Domínio Específico).
Ariel Antony Afonso - Towards a Unified Framework to Deal with Database Schema Changes in Continuous Deployment.
Master’s Students
Manoel Victor Florencio de Souza - Semantic Join Discovery in Data Lakes (Descoberta de Junções e Data Lakes com base em Anotações Semânticas).
Giovanna Andrade Santos - A Study on Sentence Segmentation in Brazilian Legal Texts Using Language Models (Um Estudo sobre Segmentação de Sentenças Judiciais Brasileiras utilizando Modelos de Linguagem).
Gustavo Rufino Feltrin - Um Estudo sobre Métodos de Sumarização de Documentos Aplicados a Acórdãos (A Study on Document Summarization Methods for Court Decisions).
Luisa P. Novaes - Métodos para Identificação Automática do Resultado de Decisões em Acórdãos no Contexto da Justiça Brasileira (Methods for Automatic Identification of Outcomes in Court Decisions in Brazil).
[Duarte’24] Aline Duarte - Um Estudo sobre o uso de Modelos de Linguagem de Larga Escala para An´alise de Dados Acadêmicos da Pós-Graduação (A Study on the Use of Large Scale Language Models for Analysis of Postgraduate Academic Data). 🎙️Posdcast-PT 🎙️Podcast-EN

Experimental Datasets Available

Schema Matching Network Datasets. This repository contains a collection of datasets related to the database schema matching problem. The datasets can be used to evaluate and compare different schema matching techniques, particularly in scenarios involving multiple schemas. Suitable for experiments in data integration, machine learning, and networked schema reconciliation. It was originally used in Diego Rodrigues, Altigran S. da Silva: A study on machine learning techniques for the schema matching network problem. J. Braz. Comput. Soc. 27(1): 14 (2021)

Support

I gratefully acknowledge the financial and institutional support provided by the following organizations:

CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico)
CNPq plays a crucial role in promoting scientific and technological development in Brazil by funding research projects, scholarships, and fostering innovation. I am honored to be recognized as a Level 1B researcher by CNPq.
CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior)
CAPES is dedicated to improving the quality of higher education in Brazil through scholarships, grants, and support for postgraduate programs. It has also provided scholarships for several students involved in my projects.
FAPEAM (Fundação de Amparo à Pesquisa do Estado do Amazonas)
FAPEAM provides financial support for scientific research and innovation within the State of Amazonas, fostering regional development. It has supported many projects throughout the years and also provided scholarships for several students involved in my projects.
Jusbrasil
Jusbrasil is a leading Brazilian legal technology company. Jusbrasil has supported a specific project under my coordination and research grants to several of my students and postdocs.

Academic Service

I am currently:

A board member of the SBC (Brazilian Computer Society)
Deputy Coordinator for Professional Programs in Computing at CAPES
Co-Editor-in-Chief of the Journal of the Brazilian Computer Society
A member of the Commission for Monitoring and Evaluation of the RNP/MCTI Management Contract.

Contact

alti@icomp.ufam.edu.br
Lattes CV ORCID
Linkedin Youtube Twitter/X Instagram
Telegram: @altigran Discord: @altigran
Office Address:
- Instituto de Computação - Setor Norte - Campus UFAM
- Av. Gal. Rodrigo Octávio, 3000 - Japiim, Manaus, AM, Brazil