Communities in PRCR
Select a community to browse its collections.
- Colecciones de la Albizu University
- Colecciones digitales de Atenas College
- Colecciones de los Centros Sor Isolina Ferré
- Materiales utilizados para la certificación.
Recent Submissions
From Catalog Fragmentation to Procurement Visibility: An Agentic Reference Architecture for Puerto Rico's First Government Item Master
(Polytechnic University of Puerto Rico, 2026) Villanueva Valentín, Jared; Torres Batista, Nelliud D.
The General Services Administration (ASG) is Puerto Rico's central procurement agency, yet its catalog contains hundreds of thousands of records with extensive duplication: a single blue ballpoint pen appears in over a hundred textual variations. This paper presents the design and implementation of an agentic Big Data infrastructure that converts the fragmented catalog into a controlled single source of truth. The architecture combines a Databricks Medallion Lakehouse with Microsoft Foundry agents that produce standardized product names and assign National Institute of Governmental Purchasing (NIGP) codes. Delivered artifacts include a daily ingestion task, cleansing and deduplication logic, five Foundry agents with responsible-AI safeguards, a unique item identifier, and a Minimum Viable Data template. The as-deployed first-year cost (~$72,000 USD) is roughly an order of magnitude below earlier hand-curation quotes. The result is a governance-aligned reference architecture for public-sector catalog modernization.
Modeling Health Risk Probabilities of Obesity and Hypertension Using Machine Learning
(Polytechnic University of Puerto Rico, 2026) Moreno Ríos, Ramón; Torres Batista, Nelliud D.
This study investigates the use of machine learning techniques to model health risk probabilities using data from the National Health and Nutrition Examination Survey 2017–2018. The objective is to evaluate the predictive performance of multiple models across health outcomes, including hypertension, stroke, kidney disease, and sleep disorders, with a focus on the relationship between obesity and hypertension. Variables such as body mass index, waist circumference, age, gender, and blood pressure measurements were used to develop predictive models. The results indicate that hypertension prediction achieved the highest performance, while stroke and kidney disease showed moderate accuracy. Sleep-related outcomes showed reduced predictive power due to limited feature representation. Additionally, the analysis confirmed a strong relationship between obesity and hypertension, as individuals with higher body mass index exhibited increased blood pressure levels. Overall, the findings highlight the potential of machine learning in health risk prediction while emphasizing the importance of data quality and feature selection.
Bilingual Hybrid Judicial-Opinion Retrieval for Puerto Rico and Texas: Dense, Lexical, and LLM-Augmented
(Polytechnic University of Puerto Rico, 2026) Pérez Cabezas, Andrés; Rodríguez Espinosa, Lisabel
This article presents the design, implementation, and evaluation of a bilingual English/Spanish information retrieval prototype over judicial opinions from Puerto Rico and Texas courts. Judicial opinions are ingested from the CourtListener REST API, cleaned, segmented into paragraph-aware chunks, embedded with a multilingual sentence-transformer model, and indexed in Facebook AI Similarity Search (FAISS) for dense retrieval, with a parallel Best Matching 25 (BM25) index supporting lexical retrieval. A weighted Reciprocal Rank Fusion (RRF) layer combines semantic and lexical rankings. An optional layer adds query rewriting and cross-encoder reranking using a locally hosted Gemma 3 large language model. On a stratified held-out split of thirty-two queries, weighted hybrid RRF reaches a normalized discounted cumulative gain at rank 10 (nDCG@10) of 0.5615, while cross-encoder reranking raises this value to 0.7680 at roughly twenty times higher latency. The system preserves end-to-end traceability to source judicial opinions and is presented as a research prototype, not a legal-advice product.
Knowledge Discovery for HR Turnover Risk and Financial Exposure
(Polytechnic University of Puerto Rico, 2026) Medina Rosario, Derek I.; Torres Batista, Nelliud D.
Employee turnover creates financial and operational challenges through recruiting, onboarding, training, productivity loss, and reduced workforce stability. This project applied a simplified Knowledge Discovery approach to estimate employee turnover risk and financial exposure using workforce indicators. A synthetic dataset of 15 employee records was created to simulate department, monthly salary, overtime status, job satisfaction, and years at company. Excel was used to transform the dataset by calculating turnover risk score, turnover risk class, retention priority segment, annual salary, and estimated financial exposure. Power BI was then used to create a basic data model, data analysis expressions measures, and an interactive dashboard. The results identified high-risk employee groups, department-level financial exposure, and workforce risk patterns related to overtime, job satisfaction, and tenure. The project demonstrates how simple workforce data can be transformed into decision-support insights for human resources planning and retention strategies.
“How to”, Practical approach to Multi Agent Cyber Defense
(Polytechnic University of Puerto Rico, 2026) Pérez Narváez, Dylan Renier; Rodríguez Espinosa, Lisabel
This project is a guided "how to" approach that explains how multi agent defense is studied, how theory guides exploration, and how practical understanding develops. It includes two hands-on proof of concepts, one built in n8n and one in Python, that put these ideas into practice using real tools. The goal is to help connect research and application in order to simplify the study of the field. This will give the reader both conceptual understanding as well as a real application of the material.