Authors - Shital Khaparde, Rakesh Verma Abstract - In the field of Speech Emotion Recognition (SER) the research is gaining more attention, with the ability to augment human-computer interaction by allowing systems to recognize emotional states from speech. The research area has evolved significantly, with most of the advancements being due to deep models of learning which are particularly good at extracting subtle patterns from raw audio data directly. The review discussed the current research, with focus on the use of prevalent deep learning architectures—e.g., Long Short-Term Memory networks (LSTM), Recurrent Neural Networks (RNN), Convolutional Neural Networks (CNN), , and hybrid models like CNN-BiLSTM—for SER tasks. The area of research has come a long way, some of the major challenges still persist. The design of dependable SER systems is still beset by issues such as imbalanced datasets, the ongoing difficulty of effective feature selection, noise sensitivity, and computational efficiency problems. Although deep learning has advanced remarkably, the review emphasizes that significant obstacles must still be overcome before the area can provide consistently trustworthy performance in actual situations. This paper proposes several future research including broadening and diversifying datasets, increasing the cross-linguistic strength of the model, and better real-time system performance.
Authors - Sharon Elizuba Koshy, Shyam A.V, Padmadas Sundaram, Sofia Rani Shaik Abstract - The growing complexity of urban ecosystems, fueled by rapid population expansion, is intensifying the challenges cities face in achieving sustainability, circularity, and resilience. This evolving landscape demands innovative management approaches that integrate technological advancements and resource optimization strategies to support resilient urban development. This research proposes a management-oriented framework for Artificial Intelligence enabled Industrial Symbiosis networks within smart cities, focusing on enhancing resource recovery and establishing closed-loop systems. By leveraging AI technologies of predictive analytics, dynamic matchmaking, and optimization algorithms, the framework aims to identify, coordinate, and scale symbiotic partnerships among industries, utilities, and municipal stakeholders. Rather than emphasizing the technical development of AI algorithms, the study addresses strategic governance models, stakeholder engagement mechanisms, and policy interventions that facilitate the effective integration of AI in managing industrial symbiosis networks. Drawing on principles of the circular economy and smart infrastructure governance, this research highlights how AI can drive adaptive and resilient resource flows within urban environments. By bridging technological innovation with strategic management practices, the study contributes to the evolving discourse on sustainable smart city development and offers actionable insights for policymakers, city planners, and infrastructure managers.
Authors - Mohar Banerjee Biswas, Srikant Das Abstract - Individuals display different levels of innovativeness towards technology and this innovativeness is a crucial personality trait to understand adoption behavior. Personal Innovativeness in Information Technology (PIIT) has emerged as a strong personality trait that influences the adoption behavior towards Information and Communication Technologies (ICT) in higher education. This research analyses the influence of PIIT on the perception of utility and simplicity of ICT by higher education students. The study extends the Technology Acceptance Model (TAM) by adding PIIT and analyses how it affects the perceived usefulness (PU), perceived ease of use (PEOU) and subsequent behavior intention (BI) towards ICT adoption. For the data collection, 493 sample data were obtained from university students using an online structured questionnaire and was processed using Structural Equation Modeling (SEM). The findings of the study reveal that PIIT significantly impacts the PU and PEOU, indicating that students who are more innovative will view ICT as beneficial to their academic task and will find it easy to use. PIIT also has an indirect influence on BI via the construct of PU and PEOU, emphasizing its mediating role. The findings propose that developing PIIT amongst learners can improve students' preparedness to adopt ICT in academics. The research has important implications for stakeholders in developing specialized interventions, training, and curriculum designs that nurture creative mindsets, thus enabling stronger integration of technology in learning.
Authors - Santanu Mandal, Srinija M, Sukruthi M Abstract - This study investigates the structural barriers hindering digital well-being among Generation Z (Gen Z) through an integrated lens of Self-Determination Theory (SDT) and Cognitive Load Theory (CLT). Using a cross-sectional survey conducted between June and July 2025, 126 valid responses were collected from Indian Gen Z participants, comprising students and early-career employees. Ten barriers—including lack of awareness, absence of digital literacy, mixed media messages, platform design addiction, and peer pressure—were examined for their interrelationships using Interpretive Structural Modelling (ISM) and MICMAC analysis. Results reveal that foundational barriers such as lack of awareness, absence of digital literacy, and mixed media messages act as high-driving factors, creating cascading effects on dependent barriers like inconsistent sleep patterns, low perceived control, and productivity guilt. From the SDT perspective, these drivers undermine autonomy and competence, while CLT explains how cognitive overload from addictive design features exacerbates maladaptive digital behaviours. The findings highlight the need for multi-stakeholder interventions—spanning policy, education, workplace culture, and ethical platform design—to address root causes and promote sustainable digital wellness. This research advances theoretical integration between SDT and CLT, offering a diagnostic framework for targeted, systemic interventions aligned with SDG 3 and SDG 12.
Authors - Rupsa Sarkar, Jaydev Mishra Abstract - Natural language processing (NLP) has several key applications, including sentiment analysis (SA). SA can bed escribed as a procedure that identifies the polarity of a sentence as well as its goal through analysis. SA is now the most active NLP research area. Aspect-based sentiment analysis, which is a subset of SA, is the process of examining a sentence's structure and determining its polarity. This field has been grown because now people feel free to share their thoughts, opinions, expressions. Internet, social media are now the massive resource of opinion assuming. In this paper, we have used Sem Eval 2014 task 4 dataset to focus on Aspect-based Sentiment Analysis (ABSA) problem. We have focused on previous work Based on Aspect-based Sentiment Analysis and Some work of Sentiment Analysis based on conversational data. We have used GPT2 for this task.
Authors - Mark Anthony C. Ochoa, Aloysius J. Aurelio, Stephan Kupsch Abstract - This sub-study evaluates the application of descriptive and inferential statistics in graduate research at DMMMSU-CGS. Findings show that while nominal and interval/ratio scales are often treated with appropriate measures (frequency counts and means), ordinal data is frequently misrepresented using the mean instead of median or mode. Such lapses compromise validity and highlight gaps in statistical literacy. This misuse underscores the need for statistical diagnostics prior to test selection and the inclusion of assumption checks in research guidelines. Moreover, analysis of appropriateness and accuracy revealed that while proper methods generally produce accurate results, some accuracy still arises from misapplied tools due to compensating factors. These findings stress the importance of enhanced statistical training and methodological rigor. . . .
Authors - Jordan Cardenas, Fabian Cardenas, Marcos Levano, Billy Peralta Abstract - In today’s digital economy, where personalization has become a cornerstone of effective marketing strategies, companies face the dual challenge of increasing advertising impact while safeguarding sensitive customer information. Despite the rapid progress of large language models (LLMs), existing commercial solutions often neglect the integration of synthetic data to reduce privacy risks and enhance adaptability, leaving organizations dependent on external providers. To address this gap, our work fine-tunes open-source LLMs (LlaMa2, Mistral, and Zephyr) with synthetic datasets generated via GPT, aiming to produce customized marketing emails tailored to demographic and behavioral features. This thesis demonstrates not only the feasibility but also the competitiveness of such models by evaluating outputs with standard metrics (BLEU, ROUGE) and human-like scoring through GPT-4, showing that open-source models can approximate the performance of proprietary alternatives at significantly lower cost. The results confirm that fine-tuned LLMs with synthetic data represent a viable solution for enterprises seeking efficiency, personalization, and internal control of data.
Authors - Rricky Tim Solomon Sison, Romary Reyes Lincod Abstract - Diversity management has become a strategic priority in the hospitality industry, particularly in regions marked by increasing demographic variation. This study investigates the perceived effects of diversity management on the work environment in Department of Tourism (DOT) accredited hotels in Western Pangasinan, Philippines. It also examines the relationship between administrative diversity management application and employees’ experiential perceptions. This quantitative-descriptive research employed an adapted and expert validated instrument, refined through pilot testing. Using purposive and stratified sampling, data were collected from 53 participants (10 supervisors/managers and 43 employees) across five DOT-accredited hotels. Four sub-variables were examined: employee engagement, creativity and innovation, stereotyping and bias, and employee retention. Analytical tools included mean, average weighted mean (AWM), one-way ANOVA, and Pearson correlation. The study revealed that diversity management was perceived to be highly effective, particularly in fostering stronger employee engagement and enhancing creativity in the work-place. Overall, its impact on the organizational environment was regarded as very positive. Nevertheless, the findings also indicated that administrative strategies showed a weak and statistically non-significant relationship with employee perceptions, suggesting that structural approaches alone may not be sufficient to shape employees’ views. Among the demographic factors considered, marital status emerged as the only variable with a meaningful influence, highlighting its relevance in understanding how diversity initiatives are experienced within the hospitality context. The study suggests a disconnect between diversity policies and their perceived impact, highlighting the need for inclusive implementation. Limitations include the small, localized sample. The findings emphasize the im-portance of participatory diversity planning, inclusive leadership training, and institutional mechanisms to ensure meaningful and sustainable diversity out-comes in hotel organizations.
Authors - Sebastian Carrasco, Pablo Schwarzenberg, Marcos Levano, Carla Taramasco, Billy Peralta Abstract - In an era where rapid, accurate keyboarding underpins virtually every academic, professional and everyday digital interaction, conventional drill-based tutors still struggle to sustain user engagement and to adapt difficulty in real time. Addressing this gap, we present Terra INVicta, a browser-native serious game that defends Earth from procedurally generated “cosmic” threats only when the player types their associated words correctly, thereby combining cognitive processing and motor skills practice in an engaging and accesible way. This paper details the game’s design workflow and a first-round evaluation of its gameplay mechanics and adaptive-difficulty engine, which modulates challenge through a mixed time–score progression factor. Initial tests indicate that Terra INVicta delivers fluid animation, near-instant keystroke-to-action responsiveness, and sustains a “flow” state across diverse player abilities, confirming its viability as a cognitive training game. Beyond validating its educational value, we highlight lessons learned and propose future work—including their application as a tool for cognitive assessment and improvement of cognitive and motor-skills in older adults. An interactive prototype is available at https://inverosimilitudes.github.io/ TerraINVicta/.
Authors - Sanjeeb Prasad Panday, Ravi Gautam, Basanta Joshi, Aman Shakya, Anunaya Pandey Abstract - The proposed approach focuses on autonomous UAV (Unmanned Aerial Vehicle) navigation and obstacle detection using only visual sensors specifically, the on-board front camera. Unlike traditional methods that rely on multiple sensors (e.g., LiDAR, radar, or GPS), this vision-based system aims to reduce hardware complexity and cost while maintaining robust performance in diverse environments. By leveraging computer vision and deep learning techniques, the UAV processes real-time camera feed data to detect obstacles, map surroundings, and plan collision-free paths. Key challenges include handling dynamic environments, varying lighting conditions, and real-time processing constraints. The system employs feature extraction, depth estimation, and semantic segmentation to interpret visual data, enabling the UAV to navigate autonomously without external aids. Advantages of this approach include reduced sensor dependency, lower power consumption, and improved adaptability in GPS denied or cluttered spaces (e.g., indoor settings or dense urban areas). However, limitations may arise in low-visibility conditions (e.g., fog or darkness) or with texture-less surfaces that complicate depth perception. The method aligns with advancements in lightweight AI models optimized for edge computing, ensuring efficient onboard processing. Future enhancements could integrate multi-camera setups or hybrid sensor fusion for increased reliability. Overall, this vision only navigation strategy offers a scalable and cost-effective solution for UAV autonomy, particularly in applications like surveillance, inspection, and disaster response where simplicity and agility are critical.
