UPLB ICS Peak One

This project introduces the development of the Behavioral and Experimental Economics Research (BeEER) application—a web-based tool designed to facilitate the teaching and execution of behavioral economics experiments in academic settings. Using the Agile methodology, the development process was divided into two iterative stages: an Alpha Version and a Beta Version. The Alpha Version was deployed in a live classroom setting with ECON 134 students to evaluate its usability and func- tionality. Results showed a strong System Usability Scale (SUS) score of 81.85 and insights into participant engagement; however, networking issues led to incomplete data transmission. To address these limitations, a Beta Version was developed using the oTree framework with a two-server architecture. This version achieved 100% data completion across all five experimental games, demonstrating improved participant ratio. The professor noted that the BeEER application is a valuable contribution to the ECON 134 course, highlighting its flexibil- ity, user-friendliness, and seamless integration with the oTree platform in enhancing the teaching and learning of economic concepts.

The exponentially growing scale of genomic datasets calls for efficient data storage and retrieval systems, partic- ularly for platforms like SNP-Seek. This study evaluated six genotype compression tools—GSC, GBC, XSI, Savvy, BCFTools, and PLINK2—based on compression ratio, compression speed, and extraction query performance using multiple SNP datasets. The query types replicated common SNP-Seek operations and were tested uniformly across all tools. The results revealed that no single tool dominated all evaluation categories. However, PLINK2 emerged as the overall best-performing tool, offering the fastest compression speeds, the best SNP list-based query responsiveness, and consistent reliability across dataset sizes. A standalone wrapper application was developed using PLINK2 to replicate selected SNP-Seek query functionalities, demonstrating its suitability for integration into web-based platforms. Compar- ative analysis with SNP-Seek’s current implementation revealed several architectural limitations, including reliance on network latency and unoptimized list-based query handling. PLINK2 consistently outperformed the current backend in most query types, particularly for smaller datasets and list-based queries. These findings confirm PLINK2’s potential to enhance SNP- Seek’s scalability, responsiveness, and storage efficiency in real- world genomic platforms. For future work, further investigation is needed to assess the full impact of PLINK2 when integrated into SNP-Seek’s production infrastructure. This will support a more fair and detailed comparison under identical system-level conditions.

At the University of the Philippines Los Baños (UPLB), students face significant challenges in managing their academic programs efficiently due to outdated systems, technical issues, and complex requirements for course planning. This study developed a web-based coursework planning system to streamline the creation, validation, and tracking of academic plans for UPLB students. Built using the PERN stack (PostgreSQL, Ex- press.js, React.js, Node.js), the system features modules for course planning, progress tracking, and a searchable course database. By integrating data from UPLB’s DX Academic Management Information System (AMIS), the tool allows students to plan their academic journey effectively and avoid common registration pit- falls. A mixed-methods approach, including the System Usability Scale (SUS) and qualitative feedback, was employed to evaluate the system’s usability and effectiveness. The system achieved a SUS score of 87.76, placing it in the “Excellent” to “Best” usability range, indicating a highly positive user experience.

The tech industry has grown to be more and more volatile as time progresses. It is generally understood that the tech industry flourished during the COVID-19 pandemic. The combination of increasing dependence on digital products and software companies’ suitability to the work from home arrangement resulted in the tech industry booming in terms of opportunities. However, this post-pandemic setting has become an adjustment period where companies reevaluate their business strategies and company structures. This has resulted in massive layoffs and outsourcing, which have only grown to be more prevalent during this reactionary period. As it stands, this change in dynamics has shook the balance in the tech industry. This, along with heightened global interest and the rise of AI and automation poses a big problem for entry-level tech professionals who will now be competing in a very saturated field. As such, this study leveraged the power of machine learning, Support Vector Machines in particular, to create models that will predict the employability of Computer Science undergraduates, in terms of their job title and time to employment. The resulting models were able to achieve satisfactory results in terms of having performance improvements over baseline models. Specifically, the classification model achieved a 58.9238% performance gain while the regression model achieved a 14.7275% performance gain. The study also developed a web application that will integrate the resulting models so as to provide an easy to use interface for would-be users. System usability testing showed that the created web application had excellent usability. Overall, this study was successful in creating predictive models for computer science undergraduates’ job title and time to employment, as well as a web application that integrates said models.

Privately owned bicycles on campus often become underutilized, remaining idle and occupying space in dormitory hallways. Elbisikleta is a mobile application developed to serve as a platform for an on-campus bike rental system based on the principles of a sharing economy. This study presents the design, development, and evaluation of Elbisikleta, utilizing Flutter as the frontend framework and Firebase as the backend service provider. The usability of the mobile application was evaluated using the System Usability Scale (SUS). The testing resulted in an average score of 88.4. This high score indicates strong user satisfaction and suggests that Elbisikleta has the potential to serve as an on-campus bike rental platform and make bicycles become a more accessible mode of transportation for students.