Central Philippines State University

Since it makes it possible to automatically analyse and produce human language to support institutional, administrative, and instructional processes, natural language processing (NLP) has emerged as a key element of artificial intelligence applications in education. NLP-based initiatives have expanded in the Philippine context in recent years, but the studies that are currently available are still dispersed and mostly application-specific. 22 relevant studies were systematically chosen and examined from a thematic review of peer-reviewed research that was retrieved from various electronic repositories. The review focused on publications from 2019 to 2025. The review reveals that the majority of NLP adoption occurs in higher education institutions, where its applications are frequently found in conversational agents, automated writing assessment and support, sentiment analysis and feedback interpretation, and institutional analytics. The synthesis indicates a shift from isolated instructional tools and toward integrated, language-driven systems that make institutions more responsive and support data-driven decision-making. Even with these improvements, there are still gaps in the use of NLP in basic education and in the discussion of ethical and governance issues. This review shows how NLP could be a strategic tool for enabling smarter educational institutions, but it also stresses the need for responsible, inclusive, and contextaware use in the Philippine education system.

The safety of school networks has become increasingly critical in today’s digitally connected world due to the growing frequency and sophistication of cyber threats such as malware, viruses, and unauthorized intrusions. Educational institutions are particularly vulnerable because of their expanding reliance on digital platforms for administrative functions, online learning, and internal communications. In many universities, a significant challenge lies in the absence of efficient mechanisms to detect and respond to cyberattacks in their early stages often identifying breaches only after substantial damage has occurred. This research introduces a unified, tri-layered framework that combines real-time threat detection using the Random Forest algorithm, intelligent alerting for immediate incident response, and content filtering for policy enforcement specifically tailored to the operational and behavioral dynamics of educational networks. By utilizing real-time traffic data from the university’s own infrastructure, the system is trained to detect a wide range of threat patterns while reducing overfitting, a common challenge in cybersecurity applications. The framework represents a shift from traditional reactive models to a proactive, predictive security paradigm, enhancing the capability of Management Information Systems (MIS) teams to preemptively address cyber risks. This integration of technical robustness, contextual relevance, and administrative control renders the proposed system both practically effective and strategically significant. The outcomes of this study have important implications for the field of educational cybersecurity, offering a scalable and adaptable model that can serve as a foundation for future research and implementation across academic institutions.

The study aimed to find out the efficacy of translation as a preliminary activity to enhance Grade 7 students’ comprehension of Filipino literature during School Year 2025–2026. Two groups participated the study, namely the control and the experimental group. A researcher-made questionnaire was used to determine the students’ comprehension of Filipino literature. The data were analyzed using mean, standard deviation, Mann–Whitney U test, and Analysis of Covariance (ANCOVA). The pre-test results of both groups indicated that they were at the “Approaching Proficient” level. In the post-test, the experimental group already achieved the “Proficient” level, whereas the control group only reached the “Approaching Proficient” level. The results of the study showed that translation as a learning strategy can significantly enhance students’ comprehension of Filipino literature. The study also supported the integration of translation as a preliminary activity in instruction.

Sugarcane productivity in the Philippines is threatened by ringspot disease caused by Epicoccum sorghinum. This study evaluated the antagonistic potential of sugarcane endophytic bacteria against E. sorghinum using Dual Culture (DCA) and Volatile Compound Assays (VCA). Molecular identification via 16S rRNA sequencing confirmed the bacterial identities. Burkholderia gladioli exhibited the highest inhibition in DCA (57.79%), while Bacillus zhangzhouensis was most effective in VCA (49.56%). Stenotrophomonas rhizophila also demonstrated inhibitory activity (16.55%). These results indicate that these endophytic strains are promising, sustainable biocontrol alternatives to chemical pesticides for managing sugarcane ringspot disease. Future work should focus on validation in screenhouse and field testing.

This research focused on applying a list of machine learning models to identify the best Substrate to increase the growth of a mushroom. In this study, three (3) distinct types of substrates were applied and analyzed. Substrate A (79 % sawdust, 1 % lime, 5 % molasses, 15 % rice bran), Substrate B (63 % rice straw, 10 % vermicast, 27 % decomposed sawdust), and Substrate C (50% rice straw, 30% sawdust, 20% forest topsoil). A total of 90 samples were cultivated and evaluated during the first harvest, detailing the growth performance of the mushroom in terms of weight, Height, and the number of stems. The growth outcomes from the collected samples were categorized using classification models, including decision trees, logistic regression, random forest, k-nearest neighbors, and support vector machines. The findings show that most models achieved high accuracy, ranging from 89 % to 100 %, which effectively distinguished them based on growth patterns. Among regression models, both yielded the best predictive performance with an ${R}^{2}$ value of 0.1. Results revealed that Substrate B produced the highest mushroom yield, attributed to its balanced composition and nutrient-rich properties from vermicast and decomposed organic matter. The study demonstrates that applying machine learning models with mushroom cultivation can enhance substrate optimization and yield forecasting. Lastly, this study would contribute to the agribusiness and agricultural sectors by developing new techniques for evaluating different substrate formulations, showcasing how machine learning can support sustainable, evidence-based approaches in agricultural biotechnology.