Bohol Island State University

Reliable, plot-scale information on irrigation status is vital to farm productivity and climate resilience. However, satellite products face limitations from cloud cover and coarse resolution, and sensor deployments can be costly in rural settings. This study presents a lightweight deep learning classifier that operates on smartphone field photos delivered as byte-buffer images from a REST API, with optional geotemporal metadata. Using the Local Irrigation System Assessment (LISA) platform, we analyzed a cross-sectional dataset of 40 labeled rice-plot images (Wet/Dry) collected in Zamora, Bilar, Philippines. A labeling SOP was implemented (checklist for visible water cues, double annotation with adjudication on an audited subset). On a fixed, stratified hold-out (Evaluation A; 13 test images), an image-only MobileNetV2 achieved 92.31

This study explored the utilization of Sagisi Palm Bud (Heterospathe elata) as a primary ingredient in three dessert products: ice cream, jam, and sorbet. Specifically, it aimed to determine the products’ production requirements, preparation procedures, nutritional composition, shelf life, and level of acceptability in terms of appearance, aroma, taste, texture, and packaging. The study employed a quasi-experimental research design combined with descriptive sensory evaluation. A total of 100 purposively selected respondents, composed of Food Technology instructors and students, participated in the evaluation using a modified 4-point Hedonic Scale. Weighted mean, Friedman test, and Durbin-Conover pairwise comparison were used to analyze the data. Nutritional analysis was conducted through laboratory testing, while shelf-life observations were monitored over several weeks. Findings revealed that all three dessert products were generally acceptable; however, Sagisi Palm Bud jam obtained the highest overall acceptability rating (3.51), followed by ice cream (3.38), while sorbet received the lowest mean score (3.17). Jam also demonstrated the longest shelf life, remaining stable for up to 45 days under proper storage conditions, compared to 30 days for ice cream and sorbet. Nutritional analysis showed that Sagisi Palm Bud is a valuable source of carbohydrates, fiber, magnesium, phosphorus, fats, and protein. Statistical analysis further revealed significant differences among the treatments across appearance, aroma, taste, and texture, particularly favoring the jam variant. The findings suggest that Sagisi Palm Bud is a promising indigenous ingredient for dessert innovation, with strong potential for commercialization, food sustainability, and income-generating opportunities.

This study explored how socioeconomic status (SES) and parental support influence the academic performance of Grade 6 learners in the Third Sub-Congressional District of Bohol during the School Year 2025–2026. Using a descriptive-correlational design, data were gathered from 250 learners selected through simple random sampling. A researcher-adapted questionnaire was used to collect information on SES and parental support, while academic performance was based on official school records. The data were analyzed using frequency, percentage, weighted mean, Pearson correlation, and regression analysis. The findings showed that most learners came from low-income families with limited access to educational resources. Despite these challenges, parents were generally highly supportive in terms of emotional, financial, educational, and social aspects. Learners were able to maintain satisfactory academic performance, with an average grade of 86.88. Among the SES factors, parental income was the only variable found to have a significant relationship with academic performance (p < .05). In addition, parental support was found to have a moderate positive relationship with academic achievement (r = 0.501, p < .01). Overall, the study highlights that while financial limitations remain a challenge, strong parental support can make a meaningful difference in helping learners succeed academically

This study determined the beginning teachers’ experiences in the Induction Program for Beginning Teachers (IPBT) and their longitudinal performance as measured by the Individual Performance Commitment and Review Form (IPCRF), as well as the challenges encountered and their influence on performance in Congressional District 1, Bohol for School Year 2025–2026. A correlational research design was used with a structured survey questionnaire. Weighted mean, standard deviation, mean, Pearson Product-Moment Correlation Coefficient, and linear regression analysis were employed to analyze the data. The findings revealed that beginning teachers demonstrated a very high level of experience in the IPBT (M = 3.74, SD = 0.44), particularly in professionalism, personal welfare, and accountability. However, they also reported a high level of challenges (M = 2.62, SD = 0.96), primarily due to a lack of resources, data gaps, and limited access to qualified mentors. While 86% of respondents achieved a very satisfactory performance level, statistical analysis showed no significant relationship between IPBT experience and performance (r = -0.09, p = 0.396). Furthermore, regression analysis confirmed that challenges encountered do not significantly intervene in this relationship (R² = 0.00865).It is concluded that while beginning teachers are highly engaged in the IPBT and perform satisfactorily, their experiences and challenges do not significantly affect performance outcomes. Other contextual factors may play a greater role. It is recommended that schools strengthen mentoring, improve resources, enhance coordination systems, and conduct further studies on other determinants of teacher performance.

This study employs logistic regression to evaluate organizational readiness within the 7S framework (Shine, Sort, Straighten, Standardize, Sustain, Spirit, Safety) using preliminary audit data from BISU–Candijay Campus. The model classifies institutional components (e.g., “Doors,” “Emergency/Security,” “Parking Area,” “Altar”) and is complemented by coefficient-based feature-importance analysis to identify offices most associated with predicted classes. Classes with sufficient representation (e.g., “Altar”) achieve comparatively higher precision/recall, whereas rare or absent classes in the test split yield near-zero scores, reflecting sample imbalance and coverage gaps. A single 80/20 holdout evaluation indicates stable but modest generalization (overall accuracy ≈ 58