RESEARCH EXHIBITS
Explore innovative research projects from 4th year BSIT JRU Students
Pre-Disaster Response and Emergency Planning System (PREP): Predictive and Prescriptive Analytics for Flood and Typhoon Analysis in San Juan City
The Pre-Disaster Response and Emergency Planning System (PREP) is designed to enhance San Juan City’s readiness against floods and typhoons. This study addresses the critical need for a unified digital solution that strengthens disaster planning and coordination through predictive and prescriptive analytics. It aims to build a comprehensive platform that supports timely alerts, resource tracking, and informed decision-making for disaster response. The literature review explored technologies used in disaster management, predictive modeling, and emergency communication systems. Studies highlight the value of real-time data, mobile alert systems, and historical trend analysis in improving preparedness. Predictive analytics, when combined with hazard maps and weather indicators, effectively identify high-risk zones. Despite these advances, gaps remain in applying integrated solutions tailored to local disaster management needs. This review supports the relevance of PREP by applying modern data strategies to improve coordination and response in vulnerable areas. This research uses quantitative and qualitative approaches, focusing on document analysis, case reviews, and system prototyping. Data sources include disaster records, scholarly articles, and official reports. Tools such as hazard mapping software, SMS gateways, and analytics dashboards are used in system development. The methodology supports PREP’s goal of integrating technology and data for proactive disaster management.
Legal Case Management System: Applying Document Security and a Recommender System
The Legal Case Management System (LCMS) project was developed to address operational inefficiencies, fragmented workflows, and security vulnerabilities in legal case handling at Delloro & Saulog Law Offices (D&S Law). Traditional manual processes and generic cloud-based tools resulted in disorganized document storage, slow retrieval, inconsistent tracking, and increased risk to confidential client information. To modernize these operations, the LCMS integrates a secure Document Management System (DMS) with Optical Character Recognition (OCR) for full-text search, Real-Time Case Tracking, and a Machine Learning–enhanced Lawyer Recommender System powered by an XGBoost Classifier. The system also implements strict role-based access controls and encryption, ensuring the confidentiality, integrity, and availability of sensitive legal documents. A key component of the LCMS is its data analytics module, which supports both descriptive and predictive insights. Descriptive analytics summarizes historical firm data such as case volumes, lawyer workloads, and case-type frequencies to help identify trends and operational bottlenecks. Predictive analytics extends this capability by forecasting potential case delays, workload imbalances, and optimal lawyer assignments. The five-step data analysis workflow (Define, Collect, Clean, Analyze, and Interpret & Visualize) ensures that operational data from appointments, engagements, case records, lawyer performance metrics, and enriched matter summaries is transformed into accurate, meaningful insights for decision-making. The XGBoost-based Lawyer Recommender System was evaluated using accuracy, precision, recall, F1-score, and confusion matrix metrics, demonstrating strong classification performance in identifying the most suitable lawyer based on expertise, availability, and past success rates. The LCMS was evaluated using ISO/IEC 25010 software quality standards and OWASP security guidelines. Results from user testing indicated improved workflow efficiency, enhanced document searchability, stronger information security, and high user satisfaction. While the system is tailored specifically for D&S Law’s processes and case types, its modular architecture provides a scalable foundation for future enhancements such as AI-driven document summarization, automated legal drafting, and integration with judicial e-filing platforms. The project highlights the crucial role of digital transformation and data-driven intelligence in modernizing legal practice and improving the delivery of client-focused legal services. Keywords: Document Management System (DMS), ISO/IEC 25010, Lawyer Recommender System, Legal Case Management System (LCMS), OWASP, Real Time Case Tracking.
Academic Resource and Knowledge (ARK): An AI-Driven Platform for Enhanced Learning and Knowledge Management
The Academic Resource and Knowledge (ARK) system is a digital library platform powered by AI, aimed at facilitating more effective information access, resource discovery, and user engagement through the use of a variety of advanced technologies. The platform incorporates an intelligent smart search using NLP, machine learning, and semantic reasoning to understand the queries and fetch the relevant documents along with a hybrid recommendation engine that offers personalized suggestions based on semantic similarity, category, trend, popularity, and authorship. The event management module allows proposal submission, activity tracking, and automated notifications, and administrators receive simplified tools for review and publication. The system quality was measured against the ISO/IEC 25010:2023 software quality model criteria, where forty respondents rated functional suitability, reliability, usability, security, and performance efficiency at a high level, with an overall mean of 4.51 (Strongly Agree). The results show that ARK is a reliable, secure, and friendly-user platform that supports the digitization of library services and academic engagement.
BPLOOP: A Business Mapping System for Pasig City Business Permit and License Office Operations Using Predictive Analytics
This study aims to design BPLOOP, a Business Mapping and predictive analysis System in the Pasig City Business permit and licensing office Business Permit and licensing office (BPLO) to solve the operations inefficiencies brought about by the old and semi manual processes. The system is built into Geographic Information Systems (GIS) that can be used to view in real-time the business location, to track the expiration and renewal of permits automatically, and to predict the business growth trends and commercial hotspots through the ARIMA model. The system was developed according to an ISO/IEC 25010:2023 software quality standard and tested on the basis of a developmental research design and Agile approach. The findings also reveal that BPLOOP can improve the efficiency of working operations, increase monitoring of regulatory compliance, and data-driven decision-making, indicating the successful use of a GIS and predictive analytics combination in updating local government business permit and licensing functions.
HydroQuest: Virtual Museum Tour and Sentiment-Driven Engagement for Museo El Deposito
This study explores the development of a web-based virtual tour system for NHCP Museo El Deposito, in response to the increasing demand for digital accessibility within the cultural industry. As museums become more online-friendly, this research seeks to offer an interactive and immersive environment that enriches visitor learning and involvement. Literature review looks into themes including virtual tour technologies, multimedia integration, and visitor analytics based on local and foreign research to identify how digital technologies enhance learning value and accessibility. The research utilizes a descriptive-developmental design, applying purposive sampling to target museum visitors, students, and online users, with data collection tools including online surveys, embedded quizzes, and system usability feedback forms. Tools like Google Forms for questionnaire, built-in quizzes for each gallery, and guided system tours will contribute to data collection, and descriptive statistics will be used to analyze the findings to determine visitor engagement and usability. The result will inform how interactive elements enhance user experience so that Museo El Deposito can continue to be an inclusive and engaging learning platform for diverse audience.
A Predictive Analytics-Based Web System for Forecasting Disease Trends to Support Health Preparedness in Barangay Daang Bakal
The Predictive Analytics-Based Web System for Forecasting Disease Trends was developed to address the challenges faced by Barangay Daang Bakal Health Center, where preparedness is hindered by manual documentation, delayed reporting, and lack of predictive capabilities. The system integrates predictive analytics to strengthen the health center's capacity for data-driven decision-making, disease trend monitoring, and community health preparedness. Its predictive module employs Decision Tree, Random Forest, and XGBoost to forecast disease trends using historical health data, demographic information, and consultation patterns, with Random Forest yielding the most accurate results (Test R2 = 0.95, MAE = 0.70, RMSE = 1.26). The K-Fold cross-validation confirmed model stability with an optimized R2 of 0.85. Forecasts feed into real-time dashboards and automated reporting systems aligned with health center protocols, offering predictive insights under various disease categories. Complementary features include heatmap visualization, patient record management, and automated report generation. Evaluation using ISO 25010:2023 standards confirmed the system's usability, reliability, and security, demonstrating a scalable framework for community-level health preparedness with an overall composite mean of 3.86, achieving "Strongly Agree" ratings across all quality characteristics.
Industry Partnership and Job Placement System for the Guidance and Testing Office of José Rizal University
Educational institutions are increasingly adopting digital systems to boost collaboration with business sectors, as well as enhance the employability of programs' graduates. The aim of this research project is to develop an industrial partnership and job placement system for the Guidance and Testing Office at Josá Rizal University which supports the United Nations Sustainable Development Goals on decent work and partnerships. It will be a web-based, secure application that will enable staff to manage a single database of student records, maintain a listing of participating business partners, and post job openings and internships. The proposed system incorporates automatic notifications, provides administrative dashboards for staff use, and uses a content-based job matching method (TF-IDF and cosine similarity) that will assist students in finding jobs that support their actual skills. The proposed system was developed using Agile Scrum methods, and evaluated against ISO/IEC 25010 quality criteria relating to usability, security, and reliability. The final evaluation provided positive results indicating an average score of 3.76 for usability, 3.75 for security, and 3.73 for reliability. Overall, the system appears to improve efficiency, increased visibility, and greater stakeholder involvement compared to the manual and third-party systems used to place students in jobs.
Geo-Tagging System for Efficient Solid Waste Monitoring for the City Environmental Management Department
This paper presents the design, development, and evaluation of a geo‑tagging system that supports solid waste monitoring in Mandaluyong City. The system includes a web dashboard for administrators and inspectors, and a mobile application for truck drivers and barangay staff. It allows real‑time tracking of garbage trucks, logging of geo‑tagged collection events, and reporting of incidents in the field. The system also integrates a machine‑learning model to forecast upcoming waste volume so that the City Environmental Management Department (CEMD) can plan routes and resources in advance. System quality was evaluated using the ISO/IEC 25010:2023 standard through alpha and beta testing. Results show high ratings in functional suitability, usability, reliability, performance efficiency, and security, with an overall average score of about 3.6 out of 4.0. The study demonstrates that a simple but well‑designed geo‑tagging platform can support data‑driven waste management for local governments.
Smart Barangay Governance: Integration of Machine Learning for Complaint Resolution, Business Permit Management, Online Services, and Citizen Feedback
The Smart Barangay Governance: Integration of Machine Learning for Complaint Resolution, Business Permit Management, Online Services, and Citizen Feedback is a web-based platform designed to streamline administrative processes and enhance communication between barangay officials and residents of Barangay Pitogo, Taguig City. The system addresses critical gaps in manual document processing, inadequate citizen notification systems, and fragmented complaint management that previously limited service delivery efficiency. The platform integrates three primary modules: Online Document Request Module for processing barangay clearances, permits, and certifications; Complaint and Feedback Management Module with machine learning-based automatic complaint categorization using text classification and Community Announcement Board for real-time information dissemination. An administrative dashboard provides barangay officials with centralized monitoring, record management, and data-driven decision-making capabilities. The system was evaluated using the ISO 25010:2023 quality standards across nine dimensions: Functional Suitability, Performance Efficiency, Compatibility, Interaction Capability, Reliability, Security, Maintainability, Flexibility, and Safety. Evaluation results from both alpha and beta testing demonstrated consistently high ratings (WM 4.60–4.77), indicating strong system performance and user satisfaction. The study demonstrates that digital platforms can significantly improve administrative efficiency, reduce processing time, enhance transparency, and promote active citizen engagement in local governance. Future enhancements include mobile application integration, advanced predictive analytics for resource planning, and scalability for deployment across other barangays or local government units in the Philippines.
Automated Student Retention Through Real-Time Attendance and Conduct Monitoring for the Student Development Office
This research presents the design and implementation of an Automated Student Retention System with real-time attendance and conduct monitoring for the Student Development Office (SDO) of Jose Rizal University (JRU). It addresses the legal and logistical limitations of paper-based and manual processes used to track student attendance, conduct, academic performance, and participation, which often result in delays in identifying and supporting students of concern. The proposed system provides real-time monitoring of student behavior and attendance, timely notifications to faculty, guardians, and SDO staff, and a centralized platform for data entry and reporting. Using both supervised and unsupervised machine learning methods, the system enables early detection of students at risk of academic or behavioral problems and supports proactive, preventive intervention practices. Specifically, the study aims to improve documentation and notification processes, increase stakeholder engagement, and enhance student retention by minimizing situations in which faculty are unaware of student issues, thereby reducing student and family disengagement. This shift from a reactive to a proactive intervention model aligns with the SDO’s goal of educational administrative excellence and the integration of local and global best practices in educational management. The evaluation results also highlight opportunities for digital transformation in addressing retention and behavioral challenges in higher education.
MandaCare: A Public Complaint Analysis System for the Mandaluyong City Health Department
This study presents MandaCare, a web-based complaint analysis system developed for the Mandaluyong City Health Department to automate the processing, categorization, and prioritization of citizen complaints. The system integrates Natural Language Processing (NLP), machine learning classification, sentiment analysis, and prescriptive analytics to transform unstructured complaint text into actionable insights. Unsupervised clustering identified fifteen validated complaint categories, while supervised learning models were evaluated to classify complaints, with an ensemble model selected for its balanced and reliable performance. Sentiment analysis was employed to support automated priority assignment, enabling administrators to identify urgent cases efficiently. Prescriptive analytics were incorporated to generate solution-oriented recommendations based on complaint category and priority. The system was implemented as a role-based web platform featuring dashboards, report generation, and automated notifications. Evaluation using the ISO 25010:2023 quality standard yielded strong results for both citizen and administrative access, indicating high functional suitability, reliability, security, interaction capability, and flexibility. The findings demonstrate that MandaCare is an effective and scalable solution for improving transparency, efficiency, and responsiveness in public health complaint management.
DengueWatch: Data-Driven Detection System for High Morbidity Rates in a Provincial Hospital
Dengue fever is still a big health problem in Zambales. It is hard for the local healthcare system to detect and respond to dengue cases quickly. This study presents Dengue Watch, a smart tool that uses machine learning to predict likely dengue cases. It gives health workers the right information to help them act fast and make better decisions. The goal of Dengue Watch is to improve how local health officials monitor dengue and respond to outbreaks. Many studies show that machine learning helps detect and predict health problems more accurately. But there aren’t many local tools like this in Zambales. That’s why this research is important. This study uses an experimental research design and combines different methods. To check the software’s quality, we follow the ISO/IEC 25010:2023 standard. We use surveys with rating scales to see how users feel about the system and calculate average scores to understand the results. We also collect detailed feedback through interviews to learn more about user experiences and needs. The software is developed using Agile Scrum, which helps us improve the system step by step based on user feedback. In the end, this study aims to provide a useful tool that helps health units in Zambales detect and respond to dengue outbreaks faster and better than before.
JRU-PULSE: Performance- and User-Satisfaction-Linked Service Evaluation for José Rizal University
This study addresses the critical role of administrative service quality in higher education by developing JRU-PULSE, a web-based platform designed to enhance the collection, analysis, and visualization of stakeholder feedback at Jose Rizal University. It integrates advanced natural language processing techniques, including fine-tuned BERT-based models, to perform sentiment analysis on open-ended comments and predict satisfaction trends from user feedback. The system facilitates real-time feedback collection via online surveys accessible through QR codes and consolidates both quantitative ratings and qualitative responses into a centralized database. Key functionalities include automated sentiment classification, common concern extraction, trend forecasting using linear regression, and an alert mechanism for significant changes in satisfaction. The platform's reporting dashboard offers comprehensive visualizations of feedback metrics, enabling university administrators and office heads to monitor service performance and make data-driven decisions. Evaluation based on ISO/IEC 25010:2023 standards confirm the system's effectiveness in improving the responsiveness, reliability, and usability of feedback processes. By addressing longstanding challenges in manual feedback management within Philippine universities, this study contributes to digital transformation efforts that foster continuous administrative service improvement and stakeholder satisfaction. The results demonstrate that the JRU-PULSE platform successfully integrates feedback from multiple stakeholder groups, providing a unified and comprehensive view of administrative service quality. Its real-time data processing and advanced sentiment analysis enable timely identification of emerging issues, empowering administrators to make informed decisions and implement targeted improvements. The platform's predictive analytics and alert system enhance proactive management, fostering a culture of continuous service enhancement. JRU-PULSE supports institutional goals aligned with the United Nations Sustainable Development Goals, particularly in promoting quality education, innovation, and collaborative partnerships. Overall, the system delivers tangible benefits by advancing digital transformation within the university, improving stakeholder engagement, and reinforcing sustainable administrative practices in higher education.
Automated Tax Declaration and Geospatial Zoning and Analytics System for Alfonso, Cavite
The study aims to develop an Automated Tax Declaration and Geospatial Zoning & Analytics System for Alfonso, Cavite, transforming the municipality’s manual tax declaration process into a digital platform. It will address current inefficiencies by integrating geospatial mapping and role-based access control to improve zoning visualization, land classification, and coordination between the Assessor’s and Treasury Offices. Using purposive sampling, data will be collected from 25 respondents (10 Assessor’s Office, 15 Treasury Office) to identify system requirements. The system will be developed using the Modified Waterfall Model with PHP, HTML, CSS, and MySQL, and evaluated based on ISO/IEC 25010:2023 software quality standards. This project aims to enhance transparency, accountability, and efficiency in local tax administration and support national goals for digital governance. It will also examine the system’s scalability and potential integration with broader e-government frameworks, offering insights for future LGU digital transformation initiatives.
Assistance Management and Disaster Preparedness System for Senior Citizens in Barangay Daang Bakal
The growing population of senior citizens in Barangay Daang Bakal has underscored the need for an efficient system to manage the distribution of medical, financial, and social assistance. Existing manual processes have led to data inaccuracies, service delays, tracking difficulties, and the risk of lost or duplicated records. This study aimed to design and develop an Assistance Management System tailored for the barangay’s elderly residents. Utilizing a developmental research design and the System Development Life Cycle (SDLC) framework, data were collected through interviews, observations, and document reviews involving barangay officials and senior citizen beneficiaries. The system automates beneficiary registration, tracks the distribution and status of aid, and generates detailed reports for officials. Key features include a secure database, a user-friendly interface, and comprehensive tracking of assistance history per individual. The implementation of the system is expected to improve the accuracy, efficiency, and reliability of assistance delivery, reduce administrative workload, and ensure timely aid distribution. Additionally, it enhances data security and provides a centralized platform for future planning. The study concludes that adopting digital solutions in local government units can significantly strengthen the delivery of social services, particularly for vulnerable sectors like senior citizens, by promoting transparency, accountability, and operational efficiency.
Fire Risk Assessment and Incident Reporting Geographic Information System with Time Series Forecasting for the Bureau of Fire Protection in Mandaluyong City
The Fire Risk Assessment and Incident Reporting Geographic Information System with Time Series Forecasting was developed to address outdated fire risk maps and fragmented incident documentation in Mandaluyong City, where the official fire risk map has not been updated since 2018 despite rapid urban development. The system integrates a web-based GIS platform with real-time incident reporting, fire hydrant and station mapping, and barangay-level time series forecasting using SARIMAX and ARIMA models on five years of fire incident data from the Bureau of Fire Protection (BFP) Mandaluyong. Forecast outputs are translated into low, medium, and high fire risk levels per barangay and visualized on an interactive map to support planning, resource allocation, and public awareness. The system was evaluated using the ISO 25010:2023 software quality model through alpha testing with city residents and beta testing with BFP personnel, yielding overall descriptive ratings around 4.3–4.4 (Agree to Strongly Agree) for functionality, usability, performance efficiency, and security, confirming its practicality for operational deployment. Future enhancements include mobile app integration, multi-channel alerts, and IoT-based sensor data to further improve real-time fire risk monitoring.
Web and Mobile-based Field Operations Management System for San Juan City Disaster Risk Reduction and Management Office
Disaster Risk Reduction and Management operations in modern urban communities require coordination on systems concerned with real-time duty monitoring, resource allocation, and inventory monitoring. This research seeks to design an integrated web and mobile system, particularly for the San Juan City CDRRMO, to enhance operational effectiveness, optimize disaster response operations, and facilitate data-driven decision-making. Through the mobile application, field operatives can receive task assignments, submit reports, and share real-time location and status updates, while administrators can monitor personnel deployment, resource availability, and inventory levels through the web-based interface. Role-based access control ensures data security and proper authorization across different user roles. A developmental research design is employed, with the Modified Waterfall Model serving as the System Development Life Cycle. Data gathering is done through structured interviews, surveys, and guided observations in collaboration with CDRRMO personnel. The system is assessed according to ISO 25010:2023 standards, with iterative feedback obtained through Likert-scale surveys and prototype testing to ensure conformity with user requirements. The research underscores the need for localized, multifunctional platforms designed for government disaster response agencies, with the machine learning algorithm serving as the core element in augmenting real-time decision support and predictive analysis.
Outreach and Extension Services: A Decision Support System for the Community Development Office of José Rizal University
The community development Offices (CDO) often struggle with manual processes that cause delays, poor documentation, and ineffective allocation of resources. This study introduced the Outreach and Extension Services Decision Support System (OES-DSS) for the Community Development Office of José Rizal University to address these issues. The system integrated three main components: a Community Profiling module for collecting resident demographic and socioeconomic information, a Budget Monitoring module for tracking allocations and disbursements, and an Automated Certification module for issuing and verifying outreach participation certificates. Together, these modules aim to create unified platforms that promote efficiency, accountability, and transparency in program management. The system also incorporated predictive analytics through the Random Forest Decision Tree algorithm. Historical and needs assessment data from partner barangays were used to generate recommendations on which outreach programs—such as Kalikasan, Kalusugan, Kabuhayan, and Kaalaman—would be most suitable for specific communities. These predictions were displayed alongside descriptive statistics and visualizations, supporting data-driven decisions at both the administrative and barangay levels. System development followed a descriptive-developmental approach, combining document analysis, surveys, and interviews with stakeholders. Evaluation was conducted using the ISO/IEC 25010 framework, measuring qualities such as usability, functionality, and reliability. The findings showed that while the system effectively streamlined workflows and provided clear visual insights, prediction accuracy was limited. This was mainly due to the small and incomplete dataset available during testing, which reduced the reliability of the machine learning outputs. Nonetheless, the integration of profiling, financial monitoring, and documentation demonstrated significant improvement over previous manual methods. The study concludes that even with modest predictive performance, the OES-DSS can serve as a foundation for more advanced, data-driven community planning once larger datasets become available. Future work should focus on expanding data collection, retraining models for higher accuracy, and strengthening the system’s integration with real-time survey inputs.