UPLB ICS Peak One

Expresscript is a mobile application designed for healthcare providers and patients that utilizes current features and technology of mobile devices in enhancing their medical treatment course. The application allows healthcare providers to generate medical e-prescriptions that lessens the chances of medication inaccuracies and errors. The application also allows patients to automatically generate schedules from medical e- prescriptions that will aid them in complying with their medical regimen. The usability of the applcation was evaluated using the System Usability Scane (SUS) with a score of 72.75 which can be classified as above average. Feedback and suggestions from respondents showed great interest for the application.

UPLB’s security firm, the University Police Force, is tasked with protecting property and ensuring safety of the public inside campus. Patrols are deployed within campus in an effort to maintain order and deter crime, but traditional methods are prone to inefficiency and predictability. This study aims to utilize game theory as an aid in deploying patrol units. Patrols movement must be effective without creating predictable patterns. With the use of weighted probabilities, randomization, and historical data, the UPF will be capable of scheduling optimal patrols with limited resources, while avoiding predictability at the same time. Historical data will be used to determine which areas of the campus have higher priority for patrolling. The system must also be able to deploy the patrols such that adversaries find it hard to predict their next move. Upon deploying patrols to areas, the scheduling program must take into account the distance that each patrol has travelled. This scheduling approach is seen to be more efficient and robust in scheduling patrols than other traditional methods, such as pure randomization. The scheduling system is capable of maximizing patrol coverage amidst the limitations on personnel numbers, and remain resilient against predictability.

Movie posters have been a staple in the film industry for advertising and promotion. With the continuous evolution of the film industry, the art styles of movie posters will persist in changing with their concept. Through the use of convolutional neural networks (CNN), classification of movie posters by its genres can be done. This study aims to create an application that will predict the genre of a movie using its poster through the use of modified CNNs of pre-trained models. The dataset used in this study was collected from The Movie Database (TMDB) website using their API, containing 18,000 movie posters unique for each genre. The dataset was divided into 60% training set, 30% validation set, and 10% test set, per genre to ensure the equal level of learning for each genre. The Xception and VGG- 16 models were used as the base models due to their size and effectiveness in image classification. The Xception (version 2) model returned the highest macro-F1 score of 36.93% among the models used in this study. Lastly, a desktop application was created for the users to upload their own movie posters to be multilabeled and get movie recommendations based on the genres from the database, and also to search for movie titles and get movie recommendations based on the genres of the movie posters.

Sig One is a web application that utilizes QR codes in place of wet signatures or their digital copies. It serves as a layer of obfuscation against forgery of these signatures through Discretionary Access Control (DAC) and image processing. Sig One also provides additional information about the document the QR code is attached to through the use of entries and subentries, which can be accessed by scanning the corresponding generated QR code. This study presents the development, implementation, and evaluation of Sig One. The application was evaluated using the Standardized User Experience Percentile Rank Questionnaire (SUPR-Q), which yielded the following results: Usability score of 4.63 out of 5; Trust score of 4.9; Loyalty score of 4.57; Appearance score of 4.43; and overall SUPR-Q score of 4.63, indicating a favorable level of user experience.

The traditional approach to teaching Philippine History in classrooms, characterized by a lack of reinforcement, has left students inadequately informed about the subject. This study explores the potential of game-based learning facilitated by the widespread use of smartphones. A mobile game for Android devices was developed which aimed at enhancing the learning experience in Philippine History and evaluated its effectiveness and entertainment value. The development utilized tools such as Godot, GDScript, Pixelorama, Visual Studio Code 1.89, and employed Pre-Test and Post-Test analysis alongside the ENJOY scale to assess educational and entertainment quality. GDScript was used for scripting game mechanics. Results indicated no significant difference in Pre-Test and Post-Test scores among students currently enrolled in HIST 1 or KAS 1. However, students who had previously completed these courses showed a significant improvement in their scores. The ENJOY scale results suggested that students found the game moderately entertaining and engaging.