UPLB ICS Peak One

Ensuring the authenticity of a digital educational document is a must and responsibility of educational institution. UPLB-ORDeR is a system that guarantee the legitimacy of digital documents issued to UPLB students and not forged signatures of faculty members to documents. IPFS generates the document hash and uses Smart Contract to save immutable information of documents to Ethereum Blockchain. It also makes the verification process easier. System Usability Scale (SUS) is used to evaluate UPLB-ORDeR and based on mean SUS score, the application receives a Grade A or Excellent Usability.

This study presents Yomi, a mobile application that supports Japanese language learners by consolidating Optical Character Recognition, a pop-up dictionary, and a flashcard- based spaced repetition system. The app streamlines the learning process by enabling users to extract Japanese text from image- based media, access definitions instantly, and review unfamiliar words in one cohesive platform. Developed using Flutter with tools such as Mokuro and the Free Spaced Repetition Scheduler (FSRS), Yomi focuses on facilitating vocabulary acquisition and reading comprehension through engagement with native content. Usability testing with 89 Japanese language learners resulted in an average System Usability Scale (SUS) score of 72.47, reflecting excellent usability. Participants emphasized the app’s effectiveness, intuitive interface, and potential for self-directed, immersive language learning.

Public transportation in densely populated urban centers like Metro Manila, Philippines, faces significant chal- lenges including severe traffic congestion and a lack of integrated transit information. This paper presents ”Street Smart,” a web- based transit application developed to address these issues. The application aims to enhance commuter education by providing a platform for nearest route detection, multimodal transit options based on user preference, and fostering familiarity with available services. Utilizing a technology stack including React JS, Mapbox, and AWS Location Service, alongside transit data from sources like Sakay.ph, the application offers interactive map-based route exploration. A System Usability Scale (SUS) evaluation with 20 respondents yielded an above-average score of 71.875, indicating positive usability and user satisfaction with the developed solution.

This study presents GraffiTrace, a convolutional neural network (CNN)-based system for detecting and mapping graffiti hotspots using Google Street View images. Leveraging a Faster R-CNN architecture implemented through Detectron2, the model has been trained on a combination of global and Filipino graffiti datasets to improve context-specific detection. After hyperparameter tuning, the model achieved an AP@50 of 41.54%, AP@[.50:.95] of 19.08%, and AR@100 of 44.32% on test data from Taguig City, indicating practical utility for preliminary tagging and spatial analysis. The system includes a GUI that enables users to upload geotagged panoramas, visualize detections on an interactive map, and generate graffiti prevalence reports. This approach supports early-stage deployment for urban planning, cultural research, and policy implementation in the Philippine setting.

This mobile application is a language learning application based on the Ilokano Language Acquisition and Immersion for the New Generation (ILAING) language learning program. It contains lessons from ILAING’s Ilokano language- learning program. It also consists of interactive quizzes and stories to train and test the learners’ vocabulary. It uses the Computer-Assisted Language Learning method of instruction, specifically Spaced Repetition Models, for adaptive and effective quizzes.