Senior Director & Head Technology Strategy and Innovations, Globe Telecom Inc.,nMember, Professional Regulatory Board of Electronics Engineering, Professional Regulations Commission, Philippines.
Monday June 22, 2026 10:36am - 10:44am PST JV Del Rosario RoomAIM CONFERENCE CENTER (ACC), Manila, Philippines
Authors - Ernest Joseph M. Garcia Abstract - This study examines how visual rhetoric and algospeak function as coordinated multimodal strategies of digital camouflage in Philippine Ukay-Ukay social commerce, and how these strategies evolve under algorithmic assimilation. As social media platforms increasingly use Artificial Intelligence (AI) for content moderation, recommendation, and commercial indexing, local online sellers must adjust their communication practices to remain visible, trusted, and economically active. In this context, sellers use Anti-Design visuals such as glitch-style images, collage layouts, and cluttered designs, together with coded language like “Budol,” “6–7 Condi,” and “EA Only,” to navigate platform rules while maintaining community trust. Using a qualitative case study approach grounded in Multimodal Social Semiotics, the study analyzes six digital artifacts from Facebook and Instagram marketplaces. Each artifact is examined through machine reading (AI/SMI outputs) and human interpretation (community meaning-making), revealing a gap between algorithmic recognition and cultural understanding. Findings show that visual rhetoric produces intentional visual noise that signals authenticity and human labor, while algospeak functions as coded language that evades algorithm detection and communicates trust, identity, and group norms. As AI systems improve pattern recognition, sellers adapt through emoji substitution, brand masking, and participatory formats such as “W or L” posts. However, a persistent intelligence gap remains between machine interpretation and human cultural meaning. The study concludes that digital camouflage is an integrated multimodal system shaped by human creativity and algorithmic governance, highlighting the continuing importance of cultural knowledge in digital commerce.
Authors - ES Sithpahan, NH Wanigasingha, MKA Ariyaratne, PRS De Silva Abstract - This study presents the design and implementation of an interactive coin classification and saving assistant aimed at enhancing financial literacy in children. The system combines computer vision and machine learning techniques to automatically identify Sri Lankan coins using a custom image dataset. A convolutional neural network (CNN) was developed and evaluated using key performance metrics, including accuracy, precision, recall, F1-score, and confusion matrices, to ensure robustness and reliability. The software was integrated with hardware components comprising a Raspberry Pi, touchscreen interface, servo motor, and webcam, forming a tangible coin-till device for user interaction. The end-to-end system, from coin insertion and classification to actuation and feedback, was validated through both quantitative and qualitative testing. Quantitative evaluation focused on model performance, while qualitative analysis assessed usability, engagement, and educational effectiveness based on feedback from children and parents. Results indicate high usability and engagement, with participants demonstrating increased interest in saving behavior. The study highlights the feasibility of combining AI and embedded systems to deliver educational experiences for children.
Authors - Emily Slanina, Marc Fernandes, Manfred Rossle, Daniel Gartner, Christian Koot Abstract - Video conferencing systems are essential to modern university infrastructure. Beyond functionality, their selection involves critical questions of digital sovereignty, institutional control over personal data, and dependencies on external providers. This paper combines a conceptual analysis of digital sovereignty with a qualitative case study at Aalen University. Based on staff interviews, technical documentation, and literature, it provides a structured comparison of Zoom and BigBlueButton (BBB) across functional, legal, organizational, and infrastructural dimensions. Zoom is perceived as highly reliable and user-friendly, yet its proprietary cloud-based model and non-European legal framework limit transparency and institutional control. While Zoom has introduced sovereignty-related configurations to meet European requirements, its governance remains external. Conversely, the open-source platform BBB offers superior transparency and control over data processing but requires significant institutional resources and technical expertise. Both platforms fulfill comparable functional requirements; however, they differ fundamentally in governance, hosting flexibility, and institutional influence. Selecting a system requires balancing structural autonomy with practical usability. Digital sovereignty is a context-dependent condition shaped by technical and legal frameworks rather than a static platform feature. BBB serves as a sovereignty-promoting alternative, provided that institutions possess the necessary technical resources and the willingness to manage structural organizational change.
Authors - Renzo Giovanka Rasyidin, Nicho Ras, Tiurida Lily Anita Abstract - Electronic menus are increasingly used in restaurants to improve ordering convenience, communication clarity, and service efficiency. In urban and culturally diverse markets, multilingual language options within electronic menus can support better customer understanding and reduce communication barriers during the ordering process. However, the factors that drive public acceptance and continued use of multilingual electronic menus remain underexplored. This study examines how perceived interaction quality and perceived ease of use influence trust in multilingual electronic menus, and how trust subsequently affects perceived service efficiency, customer satisfaction, and continued use intention. A quantitative design was employed using survey data from 230 restaurant customers in Jakarta who had prior experience using digital ordering systems. The data were analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM). The findings reveal that both perceived interaction quality and perceived ease of use significantly enhance trust. Trust, in turn, positively influences perceived service efficiency and customer satisfaction, while customer satisfaction is the strongest predictor of continued use intention. The model shows substantial explanatory power across the endogenous constructs. This study contributes to the literature by extending research on self-service technology and multilingual service interfaces in restaurant settings and by identifying trust as a central mechanism linking electronic menu interaction to post-adoption behavior. The findings also provide practical insights for hospitality operators seeking to design reliable, user-friendly, and multilingual electronic menu interfaces.
Authors - Tiurida Lily Anita, Farah Levyta, Teguh Prasandy, Mohd. Nor Shahizan Ali Abstract - This necessitates accelerating the digital transformation of remote marine destinations under pressure to maintain ecological quality and community value. Yet, over the course of implementing digitalization, many such destinations continue to face issues like disjointed stakeholder coordination, heterogeneous technological preparedness and lack of empirical research on sustainability-oriented evaluations of digital tourism frameworks. This research tests the interrelationships of Smart Tourism Readiness, Perceived Sustainability Value, Perceived Usefulness, Adoption Intention, and Perceived Destination Competitiveness. Using partial least squares structural equation modelling (PLS-SEM), quantitative surveys of 180 local tourism stakeholders were analysed. The results indicate that Smart Tourism Readiness gives a positive effect on Perceived Usefulness, as well as Perceived Sustainability Value, where the latter has a superior influence. Conversely, the results show that Adoption Intention is strongly predicted by Perceived Usefulness, and that both Adoption Intention and Perceived Sustainability Value significantly contribute to making Perceived Destination Competitiveness a more appealing choice. The findings show that the model has significant explanatory and predictive power regarding endogenous constructs. The findings indicate that stakeholders only tend to support digital transformation initiatives if they are perceived as practical, relevant and aligned with local sustainability priorities. How this paper contributes · The study extends the theory of technology adoption, sustainability value and destination competitiveness in a holistic empirical framework. The findings provide strategic insights to managers on how to implement digitally driven sustainable tourism development.
Authors - Mangapuram Sadhguna Sri, Narendra VG, Shiva Prasad Gundibail Abstract - This project tackles the inefficiency and possibility for bias in traditional resume screening. When a single job post draws hundreds of applications, manual review becomes a bottleneck. We designed an intelligent system that uses NLP and machine learning to automate this process. Our system features a complete pipeline that parses PDF resumes and converts their text into structured features using TF-IDF or Sentence Embeddings. A trained classifier then evaluates these features to predict the optimal job role for each candidate. By ranking and classifying applicants based on skill relevance, our tool allows hiring managers to bypass manual sorting and focus directly on the most promising individuals, ensuring a faster and more equitable screening process.
Authors - Karen Chelentine Natalia, Trias Septyoari Putranto Abstract - Even though this is now a necessity in the sector, particularly because of COVID-19,cleanliness has taken on an even greater concern that reflects itself with the implementation of Artificial Intelligence (AI) for automated hygiene monitoring throughout hospitality environments. Yet, numerous Al models re-main black boxes and their decision-making process is opaque to operators - a dynamic that can undermine trust in the technology. Bhalearo et al.: Evaluation of Explainable Artificial Intelligence (XAI) in Cleanliness Detection Systems: A Study on Hospitality Services This study used a qualitative case study method with in-depth interviews of four participants: a manager, supervisor, staff member and hotel guest as data collector. The findings suggest Al can aid efficiency and consistency in cleanliness monitoring, but that limited interpretability could also hamper their trustworthiness. XAI is an excellent way to increasing the understanding of system outputs by users, and this leads to increased trust and ac-countability through a better explanation - especially with visual explanations like heatmaps. We highlight the benefits of technological efficiency but also transparency, especially in hospitality management.
Authors - Eisenhower D. Aldemita II, Christopher Luis M. Senatin, Wilson M. Tan Abstract - As the Internet of Things (IoT) scales up to billions of connections, securing resource-constrained devices remains a challenge. The lack of lightweight trust management systems and weak access control are some of the existing vulnerabilities in IoT systems. This paper proposes a Public Key Infrastructure architecture integrating a digital national ID system to tie human and machine identity together. The system requires users to present their ID for verification with the digital national ID system before allowing registration of their IoT devices, tying their identity to the certificates issued to their devices. Ownership challenges are implemented during the registration of devices to ensure the user has physical access to and possession of the device. Once registered, mutual Transport Layer Security is used for secure communication between devices and servers. The system was evaluated on ESP8266, Raspberry Pi Pico W, and ESP 32 microcontrollers with 10000 tests per key performance metric. The results show that the proposed architecture provides increased security with minimal impact on device performance.
Authors - Lanz Anjelo Conanan, James Ernest Geraldo, Wilson M. Tan Abstract - With the increasing push for digital transformation and egovernance in order to improve service delivery capacity and crisis handling, some Philippine Local Government Units (LGU) have begun to develop their own Digital Identity Systems (DIS) as the foundation of their digital infrastructure. Due to the recognition of LGUs’ constraints in developing their own DIS and the interoperability opportunities with the Philippine Identification System (PhilSys), the mature National DIS of the Philippines, this paper developed OpenLGU ID, a customizable and open source LGU DIS which integrates with PhilSys. Its architecture implements a variety of recognized LGUs needs such as ID issuance, service claiming, and sector group management. It is designed with standards like Digital Public Goods (DPG) in mind given it’s minimal storage of Personally Identifiable Information (PII) and it’s signing of QRs. Ultimately, the researchers present an open source system ready to be implemented by LGUs to assist them in their digital transformation.
Authors - Gokaramaiah Thota, Sathya Babu Korra, Nagaraju K, Suman Prakash, Perumalla, V Ramanjaneyulu Yannam Abstract - Content-based image retrieval (CBIR) has become an important research area in computer vision for retrieving visually similar images from large-scale image collections using intrinsic image content rather than manually assigned annotations. Although deep feature learning has substantially improved retrieval accuracy, existing retrieval systems continue to encounter challenges, including limited interpretability, high computational complexity during similarity search, and insufficient integration of semantic attention information into retrieval and indexing mechanisms. Most retrieval acceleration techniques operate primarily in feature space and often neglect region-level semantic cues that contribute to retrieval relevance. This work presents a hybrid framework that integrates complete optimization of convolutional features, explainable artificial intelligence (XAI)-guided feature representation, and histogramdriven indexing to improve retrieval effectiveness while reducing computational cost. The complete convolutional neural network architecture is optimized to learn domain-adaptive visual representations, while class activation information is employed to identify semantically important image regions and generate activation-aware feature embeddings that preserve discriminative visual characteristics. The extracted features are normalized and these are subsequently organized into histogram partitions to reduce the retrieval search space. This indexing strategy limits similarity computation to semantically related candidate regions while preserving retrieval quality. Experiments conducted on Corel-10K and Oxford-5K evaluate retrieval performance using Precision@K, average precision, mean average precision, retrieval time, and average feature comparisons. Experimental results demonstrate that the proposed framework reduces the retrieval search space by approximately 35 to 55% while maintaining competitive retrieval accuracy. Comparative analysis shows competitive retrieval performance with lower computational complexity and improved interpretability.
Authors - Rajendra Yadav, Pavan Kumar Boori, Gaurav Kumawat, Chirag Joshi Abstract - Examining skin images from dermatoscopes can be a challenging endeavor especially when unlabelled skin images are being used. Therefore, we propose a novel method based on deep learning called Contrastive Clustering Autoencoder (CCA), specifically for clustering skin lesions without using la-belled data. CCA employs contrastive learning as well as clean autoencoder architecture based off a truncated ResNet-18 architecture. Internally, these models separate features into two branches: one for reconstruction of the image and one for clustering similar lesions, allowing it to learn informative patterns while maintaining image quality. Contrastive loss is used to create tight clusters of similar images and clean boundaries between them. To further increase the results, a pseudo-labelling approach is employed to take the most confident model predictions and use them to improve the model, modelling both unsupervised and semi-supervised learning methods simultaneously. CCA is evaluated against the HAM10000 dataset measured in cluster purity, silhouette score, and expert re-view. The results demonstrate that CCA can cluster skin lesions with high accuracy and consistency without the need for labeled data. The model's stability, the level of confidence in the results, and the expectations in the medical field indicate that CCA has tremendous potential for skin diagnosis-related applications where labelled data cannot be obtained.
Authors - Angeline B. Elegio, Darrel A. Cardana, Kathlyn L. Quion Abstract - This study assesses the implementation of the Intellectual Property and Technology Business Management (IPTBM) framework at Bohol Island State University (BISU), a geographically peripheral state university in the Philippines. Using a longitudinal descriptive case study design, the study reviewed institutional documents, policies, IP records, training reports, curricular materials, and technology utilization records from 2019 to 2025, with emphasis on the IPTBM implementation period from 2023 to 2025. Findings show that IP applications increased from 89 in 2023 to 156 in 2025, representing a 75.3% overall increase, while IP registrations rose from 45 to 112, equivalent to a 148.9% increase. Sectoral filings were concentrated in Food with 74 filings or 24.7%, ICT with 64 filings or 21.3%, and Creative Works with 54 filings or 18.0%, reflecting the university’s regional innovation priorities. Capacity-building also expanded, with trained participants increasing from 723 in 2023 to 1,839 in 2025, or a 154.4% increase. However, IP utilization remained largely extension-mediated, accounting for 26 of 29 recorded utilization activities or 89.7%, while commercialization through licensing and revenue generation remained emerging. Viewed through the Triple Helix framework, the findings suggest that IPTBM functioned as an enabling mechanism for strengthening IP governance, human capital development, and university-level innovation management. The study contributes longitudinal, institution-level evidence on how resource-constrained HEIs can move toward integrated IP protection, utilization, and innovation governance.
Authors - Daniel Dasig Jr, Rhoda Fabe, Karen S. Guillermo, Chad Vincent Florentino Abstract - Higher Education Institutions (HEIs) increasingly rely on data governance and quality management systems to enhance institutional transparency, sustainability reporting, and strategic decision-making. This study investigates the role of data governance mechanisms and quality management practices in strengthening Sustainable Development Goal (SDG) reporting among universities using institutional evidence derived from the 2025 Times Higher Education (THE) Impact Rankings submission. The study adopts a qualitative-descriptive and document analytics approach to examine governance structures, evidence management systems, stakeholder engagement processes, and quality assurance practices associated with SDG reporting. Findings reveal that structured data governance significantly enhances evidence traceability, policy alignment, reporting accuracy, and institutional accountability. Furthermore, integrated quality management frameworks improve cross-functional coordination among academic, administrative, and sustainability offices. The study proposes a Strategic Data Governance and Quality Management Framework (SDG-QMF) for HEIs to improve sustainability reporting readiness, accreditation compliance, and global ranking performance. The paper contributes to the growing discourse on digital governance, institutional analytics, and sustainability-driven quality assurance in higher education.
Authors - Angeline B. Elegio, Darrel A. Cardana, Kathlyn L. Quion, Max Angelo D. Perin Abstract - Universities are increasingly expected to translate research into usable technologies through effective intellectual property and technology business management (IPTBM). However, the pathway from intellectual property (IP) outputs to commercialization remains challenging, particularly in regional innovation programs where institutions operate under varying resources and capabilities. This study examined the IPTBM implementation experience of five higher education institutions (HEIs) in Region 7, Philippines, under the Regional Agri- Aqua Innovation System Enhancement (RAISE) Program, tracing the pathway from IP outputs to commercialization outcomes. A documentary analysis design was employed, analyzing official IPTBM records including IP filings, commercialization cases, capacity-building participation, and governance milestones. A total of 584 IP filings were documented, dominated by utility models (63.4%), with output concentrated in two institutions (CTU: 46.1%; Regional IPTBM/BISU: 34.1%). Five commercialization cases were recorded (0.86% conversion rate) generating Php 59,762.25 in documented licensing returns, predominantly through agreements with local cooperatives and MSMEs. Core training participation was standardized, but supplemental training and outreach varied substantially. Policy development stages ranged from drafted to Board of Regents- approved. Findings suggest that output volume alone does not ensure commercialization success; rather, market-facing capabilities, portfolio prioritization, and partner readiness critically shape technology transfer outcomes.
Authors - Sigit Bayu Cahyanto, Selo Sulistyo, Guntur Dharma Putra Abstract - Digital royalty distribution increasingly relies on metadata quality, ownership verification, and event-level usage records across heterogeneous digital platforms. However, inconsistencies in metadata governance and verification processes continue to reduce allocation reliability, traceability, and policy compliance in digital rights administration. This paper proposes Music Footprint, an event-level metadata governance framework that integrates metadata validation, provenance assessment, verification confidence, and royalty analytics to support transparent and auditable royalty allocation. The study employs a Design Science Research approach combined with system dynamics simulation to evaluate the influence of metadata completeness, ownership consistency, provenance integrity, and verification confidence on royalty eligibility and payout allocation over a 365-day operational horizon. The results demonstrate that metadata quality and verification consistency significantly influence allocation reliability and payout traceability. The proposed framework contributes to data governance and quality management by positioning metadata verification as a governance control mech- anism for accountable and policy-aligned digital royalty administration.
