La Consolacion University Philippines

Integrating large language model (LLM) APIs in an institutional web information system presents architectural challenges including service isolation, API fault tolerance, secured credential management, and cost-effective model routing. This paper presents service-oriented design patterns for embedding OpenAI GPT services within a Laravel-based system for community extension management, deployed at www.ceso.me. The architecture employs ten functional modules covering the full community extension lifecycle: user management with CSV bulk import and role-based access control (RBAC) across five user roles, activity management with unique entry code generation, content management system (CMS) and certification repository, and AI-powered feedback classification and automated reporting. The AI service layer uses a dual-model routing pattern: GPT-3.5-turbo (avg. latency: 0.8s, cost: P9.11/1K requests) handles dashboard insight synthesis, while GPT-4.1-mini (avg. latency: 1.4s, cost: P36.43/1K requests) performs granular sentiment classification with contextual reasoning, achieving 97% cost reduction compared to full GPT-4 deployment. Fault tolerance is implemented through exception-based fallback classification, achieving 98.7% API success rate with 1.3% graceful degradation during a 30-day observation period. Observed system performance metrics include average page load time of 1.2 seconds, database query time below 50ms, and successful handling of 500 concurrent users in load testing. Evaluation by 38 respondents using the ISO/IEC 25010 quality model yielded weighted mean scores of 4.82 (Functional Suitability), 4.55 (Performance Efficiency), 4.71 (Usability), 4.67 (Maintainability), and 4.75 (Decision Support Effectiveness). The design patterns and architecture presented are intended to be adaptable to institutional web systems seeking to integrate LLM services with production-grade reliability.

Academic integrity in higher education is a topic of choice, especially in the era of artificial intelligence (AI). Hence, institutions are making waves to establish a concrete, sustainable policy to regulate and monitor students and promote a credible academic experience. This study investigated the level of academic integrity and academic performance among tertiary students at a higher education institution in Olongapo City, Philippines. Using a descriptive-correlational research design, the study assessed the perceptions of 395 participants who were sampled via convenience sampling. The investigation employed a research tool adapted from a previous study comprising five latent variables: Honesty, Fairness, Respect, Trust, and Responsibility. This investigation was conducted during the first semester of the academic year 2025-2026. Employing both descriptive and inferential statistics using the Statistical Package for the Social Sciences (SPSS) version 23, the study revealed a high overall level of academic integrity, with honesty emerging as the highest-rated dimension. Academic performance was similarly robust, with 37.2% of the sample achieving a "Very Good" and 30.1% reaching "Excellent." Inferential analyses using t-tests and ANOVA indicated significant differences in integrity levels by sex at birth, college affiliation, and year level. At the same time, age specifically influenced perceptions of Fairness and Trust. Correlation analysis established significant positive relationships between all integrity dimensions and academic achievement. Furthermore, multiple regression analysis identified Trust as a significant unique predictor of academic performance, with honesty showing marginal significance. These findings suggest that a student's commitment to ethical standards is intrinsically linked to their scholastic success. The study concludes that educational institutions must implement tailored interventions that account for demographic diversity to cultivate a culture of Trust and Responsibility, ultimately fostering a sustainable learning environment that supports both ethical standards and academic outcomes.

The operational readiness of military reserve forces underpins national security and disaster response, yet persistent skill mismatches and manual task assignment hinder effective mobilization. This study developed a Web-Based Skill Management and Task Recommender System for Military Reserve Force Mobilization that integrates reservist profiling, skill tagging, task creation, automated recommendation, and notification delivery through a hybrid pipeline composed of (i) a rule-based filter that enforces minimum qualification constraints and (ii) a weighted multi-criteria scoring layer that ranks qualified personnel by skillset match, availability, proximity, and rank suitability. The system was built using the Agile methodology, and synthetic data modeled on the structure of real reservist profiles were used throughout development and evaluation to address confidentiality and operational security concerns. The system was evaluated by 26 respondents (11 IT experts and 15 military officers) selected through purposive sampling, using a structured questionnaire derived from the ISO/IEC 25010 Software Quality Model. The system was rated Very Acceptable across all eight quality characteristics, with weighted means ranging from 3.73 to 3.91. The hybrid approach was designed as a decision-support tool that surfaces transparent, auditable recommendations to reduce potential skill-to-task mismatches, while final deployment authority remains with qualified military personnel.

This study examined patterns of AI tool use and reliance among 425 college students, the perceived impact on academic performance, and the role of instructor guidance in shaping these behaviors. A descriptive-correlational research design was employed, utilizing a structured online questionnaire. Results indicated that the vast majority of respondents used AI tools several times a week (52%) or daily (34.4%), primarily to check grammar and spelling, generate ideas, and research topics. Most students (52.9% agree; 36.9% strongly agree) perceived AI tools as improving their learning experience, and 94.6% reported a positive impact on their academic performance. Spearman correlation analysis revealed a statistically significant positive relationship between instructor encouragement and AI usage frequency (rs = 0.1144, p < .05), and between instructor encouragement and students' intention to continue using AI tools in the future (rs = 0.1609, p < .001). However, no statistically significant relationship was found between students' belief that AI improves learning and their self-reported grades (rs = 0.0226, p > .05). Additionally, the purpose of AI use (writing, research, math, grammar) did not significantly differentiate grade levels among students. The findings underscore the widespread integration of AI into students' academic routines and the influential role instructors play in mediating responsible AI adoption.

Biofilms are organized groups of microbes surrounded by an extracellular polymeric substance (EPS) matrix. This structure helps microbes survive, creates metabolic differences, and makes them less sensitive to antimicrobial treatments. Because biofilms can block antimicrobials and help microbes adapt, they often cause chronic and recurring infections that are hard to treat with standard methods. Most current diagnostic and antimicrobial testing methods focus on free-floating (planktonic) microbes and do not reflect the complex structure and behavior of mature biofilms. This gap often leads to ongoing treatment failures and poor predictions of treatment outcomes. Recently, artificial intelligence (AI) and computational modeling have shown promise for improving biofilm research. These tools can help with automated detection, structural analysis, computational phenotyping, and predicting how biofilms will respond to treatments. This review examines current and emerging AI-based methods in biofilm biology, focusing on computational analysis, prediction of antimicrobial responses, and AI-supported antibiofilm therapies. It also discusses challenges such as dataset differences, limited real-world testing, difficulty understanding models, and a lack of models for clinically important mixed-species biofilms. Overall, this review shows how AI could help improve the accuracy, integration, and tailoring of biofilm research and antimicrobial management.