Eniedson Fabiano Pereira da Silva Júnior
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Eniedson Fabiano Pereira da Silva Júnior has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- André Luiz Firmino Alves
Federal Institute of Paraíba (IFPB), Picuí, Paraíba, Brazil
3 shared publications - Cláudio de Souza Baptista
Universidade Federal de Campina Grande (UFCG)
3 shared publications - Francisco Igor de Lima Mendes
Federal University of Campina Grande, Campina Grande, Paraíba, Brazil
1 shared publication - Fábio Lucas Meira de Souza Barbosa
State Court of Accounts of Paraíba (TCE-PB), João Pessoa, Paraíba, Brazil
1 shared publication - Luciano Barbosa
CESAR School, Recife, Pernambuco, Brazil
1 shared publication - Mateus Queiroz Cunha
Universidade Federal de Campina Grande (UFCG)
1 shared publication
Latest Publications
- 30 September 2026
- Published by World Scientific Pub Co Pte Ltd in International Journal of Semantic Computing
- p. 1-25
Hybrid Multi-Hop RAG for Intelligent Legal Information Retrieval
Public servants and auditors must often identify precise legal evidence across large collections of laws, decisions, audit reports, and technical documents. However, institutional search systems typically rely on rigid metadata or exact keyword matching, making it difficult to answer natural-language questions whose evidence is distributed across heterogeneous sources. This paper presents a retrieval-augmented generation (RAG) framework for legal corpora that combines hybrid retrieval, keyword extraction, metadata-driven filtering, and multi-hop retrieval for legal-administrative question answering. The study follows a staged design. Stage I establishes a keyword-augmented hybrid RAG baseline that integrates dense and lexical retrieval with automatic keyword extraction. Stage II extends this baseline with query decomposition, LLM-guided metadata extraction for dynamic filtering, and domain-specific cross-encoder reranking. We evaluate the framework on two corpora: a test set with more than 7000 legislative and administrative documents and a validation corpus with more than 333,000 PDFs and 10 terabytes of data across five institutional knowledge bases: legislation, decisions, audit reports, prosecutor opinions, and portal files. Results show that keyword-guided hybrid retrieval improves evidence coverage and ranking quality, while structured multi-hop control further improves retrieval coverage and answer correctness in heterogeneous, high-stakes legal corpora.
- 30 September 2026
- Published by World Scientific Pub Co Pte Ltd in International Journal of Semantic Computing
- p. 1-25
Hybrid Multi-Hop RAG for Intelligent Legal Information Retrieval
Public servants and auditors must often identify precise legal evidence across large collections of laws, decisions, audit reports, and technical documents. However, institutional search systems typically rely on rigid metadata or exact keyword matching, making it difficult to answer natural-language questions whose evidence is distributed across heterogeneous sources. This paper presents a retrieval-augmented generation (RAG) framework for legal corpora that combines hybrid retrieval, keyword extraction, metadata-driven filtering, and multi-hop retrieval for legal-administrative question answering. The study follows a staged design. Stage I establishes a keyword-augmented hybrid RAG baseline that integrates dense and lexical retrieval with automatic keyword extraction. Stage II extends this baseline with query decomposition, LLM-guided metadata extraction for dynamic filtering, and domain-specific cross-encoder reranking. We evaluate the framework on two corpora: a test set with more than 7000 legislative and administrative documents and a validation corpus with more than 333,000 PDFs and 10 terabytes of data across five institutional knowledge bases: legislation, decisions, audit reports, prosecutor opinions, and portal files. Results show that keyword-guided hybrid retrieval improves evidence coverage and ranking quality, while structured multi-hop control further improves retrieval coverage and answer correctness in heterogeneous, high-stakes legal corpora.
- Open Access
- 31 March 2025
- Published by Association for Computing Machinery (ACM)
- p. 1267-1274
Comparative Study of Lexical and Semantic Approaches in Closed-domain Product Search
- 31 March 2025
- Published by Association for Computing Machinery (ACM)
- p. 1267-1274
Open AccessComparative Study of Lexical and Semantic Approaches in Closed-domain Product Search
- Open Access
- 20 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-10
On Searching Product Catalog: Relational Database Versus Search Engine approaches
Context: Product queries are pivotal in various applications, such as e-commerce websites and price comparison tools. In particular, searching for products on invoices is an essential way to help monitor expenses from public entities. Problem: The volume of invoice data and the lack of standardization in product descriptions pose challenges for a search tool. In this context, query processing time and the relevance of results obtained are decisive factors. Solution: We investigated query results, considering metrics of product relevance in the retrieved outcomes, as well as query processing times in SQL Server and Elasticsearch. Statistical methods were employed to analyze the results and determine the most suitable search mechanism for the context. Information Systems Theory: We applied principles of Information Systems and experimental design techniques to conduct a comparative assessment of the effectiveness of Elasticsearch and SQL Server in searching for products in invoices. Method: We adopted a quantitative approach, utilizing a Design of Experiments (DoE) to compare Elasticsearch and SQL Server. In the relevance analysis, various queries were considered, varying the number of results returned. In the performance evaluation, we varied different parameters, including temporal filters, the number of terms in queries, and the popularity of the searched products. Summary of Results: Compared to SQL Server, Elasticsearch exhibited superior performance, providing more agile responses and delivering more relevant query results. Contributions and Impact in the IS Field: This research contributes by assisting in selecting product catalog search mechanisms, highlighting Elasticsearch as superior in performance, and promoting quantitative methods for evaluating technologies in information systems.
- 20 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-10
Open AccessOn Searching Product Catalog: Relational Database Versus Search Engine approaches
Context: Product queries are pivotal in various applications, such as e-commerce websites and price comparison tools. In particular, searching for products on invoices is an essential way to help monitor expenses from public entities. Problem: The volume of invoice data and the lack of standardization in product descriptions pose challenges for a search tool. In this context, query processing time and the relevance of results obtained are decisive factors. Solution: We investigated query results, considering metrics of product relevance in the retrieved outcomes, as well as query processing times in SQL Server and Elasticsearch. Statistical methods were employed to analyze the results and determine the most suitable search mechanism for the context. Information Systems Theory: We applied principles of Information Systems and experimental design techniques to conduct a comparative assessment of the effectiveness of Elasticsearch and SQL Server in searching for products in invoices. Method: We adopted a quantitative approach, utilizing a Design of Experiments (DoE) to compare Elasticsearch and SQL Server. In the relevance analysis, various queries were considered, varying the number of results returned. In the performance evaluation, we varied different parameters, including temporal filters, the number of terms in queries, and the popularity of the searched products. Summary of Results: Compared to SQL Server, Elasticsearch exhibited superior performance, providing more agile responses and delivering more relevant query results. Contributions and Impact in the IS Field: This research contributes by assisting in selecting product catalog search mechanisms, highlighting Elasticsearch as superior in performance, and promoting quantitative methods for evaluating technologies in information systems.