Authors - Lord Francis B. Navarro, Chris Jordan G. Aliac, Larmie S. Feliscuzo Abstract - This study benchmarks three Transformer-based encoder models for the sentiment classification stage of an aspect-based sentiment analysis pipeline applied to tourist reviews of the Chocolate Hills Complex in Bohol, Philippines. The work is motivated by the need for tourism analytics that remain usable under the computing constraints of Philippine local government units. A corpus of 5,885 Google Maps and TripAdvisor reviews was cleaned to 3,288 English textual reviews and transformed, through LLM-assisted silver-standard annotation, into 7,555 aspect-sentiment pairs across six tourism aspects and three sentiment classes. Three models — RoBERTa, DistilBERT, and TinyBERT — were finetuned for aspect-conditioned sentiment analysis and compared with TF-IDF baselines. Classification was evaluated on a held-out test set; deployment efficiency was tested on CPU-only hardware using latency, memory footprint, and parameter count. RoBERTa achieved the highest accuracy and macro-F1 but required substantially more memory and higher latency. TinyBERT achieved the lowest latency and memory use while maintaining usable macro-F1, making it the most deployment-practical option under the tested conditions. The results suggest that model selection for local tourism analytics should consider both predictive performance and operational feasibility.
Authors - Adeline Aulia Darsonoputri, Farah Alfanur Abstract - Indonesia’s local fashion industry has grown alongside digital marketplaces, social media, and live commerce, expanding market opportunities while increasing competition, customer switching, and digital platform dependence. Neulla, a Bandung-based Indonesian fashion brand with the concept of “Basic with a Twist,” faces the need to strengthen differentiation, customer relationships, and competitive positioning. This study formulates renewed business development strategies for Neulla using an integrated Business Model Canvas (BMC), PESTLE, Porter’s Five Forces, SWOT, and TOWS Matrix approach. A descriptive qualitative case study was conducted through in-depth interviews with internal and external informants supported by company documentation. The findings show that Neulla’s current business model has implemented the nine BMC blocks, with strengths in brand identity, digital sales channels, and product design capability. However, Neulla faces challenges related to competition, changing fashion trends, marketplace dependency, production capacity, and creative team turnover. The TOWS Matrix generated 30 alternative strategies, which were consolidated into 18 renewed strategies and classified into short-term, mid-term, and long-term priorities. These strategies were integrated into a new BMC to strengthen design differentiation, sales channels, customer engagement, internal systems, and partnerships. The proposed business model offers practical strategic direction for enhancing Neulla’s competitive positioning in Indonesia’s local fashion industry.
Authors - Josephine Florencia Chan, Anderes Gui, Riki, Huynh Trong Thua, Nguyen Minh Tuan, Chau Van Van Abstract - Losing customer in the telecommunication may lead to significant financial losses. Machine learning approaches have shown promising potential for predicting churn, but many studies still focus primarily on Accuracy, which can be misleading when using an imbalanced dataset. This study compares three ma-chine learning algorithms: Logistic Regression, Linear Support Vector Machine (SVM), and Decision Tree. The goal is to determine which algorithms prioritizes Recall. The Iranian Churn dataset was used for the experiment; this dataset con-sists of 3151 customer records with 14 behavioral and demographic attributes. This study used an 80:20 train-test split with standardized features, and model performance was evaluated based on Recall, F1-score, Precision, Specificity, and Accuracy. The Decision Tree model achieved the highest Recall, while Logistic Regression and Linear SVM showed slightly lower Recall but similar Accuracy. These results suggest that for small and structured customer datasets, simpler or appropriately constrained models may perform effectively while prioritizing the identification of churners. Model selection should consider dataset characteris-tics. Prioritizing Recall over Accuracy can also help guide effective customer retention strategies.
Authors - Thanh Hien Hoang, Thi Dieu Linh Huynh, Le Hoang Linh Chi Abstract - This study examines whether the institutional digitalization of trade procedures in importing partner countries is associated with Vietnam’s bilateral export performance. While existing studies have widely examined trade facilitation, e-commerce, and general ICT adoption, less attention has been paid to the role of partner-country digital trade readiness in shaping export market access for an export-oriented economy such as Vietnam. Using panel data on Vietnam’s exports to 27 major trading partners over the period 2013–2022, this study applies an extended gravity model incorporating the Paperless Trade Index, the Cross-border Paperless Trade Index, and the aggregate Trade Digitalization Index. The model also controls for importer GDP, Vietnam’s GDP, geographical distance, partner-country innovation capacity, and the COVID-19 period. The random-effects estimates show that paperless trade, cross-border paperless trade, and over-all trade digitalization in importing markets are positively and significantly associated with Vietnam’s export values. The findings also confirm the relevance of conventional gravity variables, with GDP showing positive associations and distance showing a negative association with exports. These results suggest that digital trade readiness in destination markets can function as an external institutional condition supporting export competitiveness. The study contributes to the literature by distinguishing between domestic paperless trade and cross-border paper-less trade in importing markets and by providing Vietnam-specific evidence on the strategic importance of interoperable digital trade procedures.
Authors - A. Muhammad Maheswara Iporennu, Siska Noviaristanti Abstract - This study formulates and prioritizes business strategies for Survei Kos Incaran by Kospace, a student accommodation property management service operating around Telkom University. The study is motivated by the increasing shift of accommodation search activities from conventional channels to digital and platform-based services, which raises the importance of information accuracy, service reliability, and field verification. The research applies a descriptive qualitative case study approach using semi-structured interviews, limited observation, internal documents, and operational data. The analytical process integrates Internal Factor Evaluation (IFE), External Factor Evaluation (EFE), Internal-External (IE) Matrix, SWOT/TOWS Matrix, and Quantitative Strategic Planning Matrix (QSPM). The results show that Kospace is positioned in the Hold and Maintain cell of the IE Matrix with an IFE score of 2.300 and an EFE score of 2.510. QSPM prioritizes the digitalization of real-time service monitoring and scheduling as the highest-ranked strategy with a total attractiveness score of 5.450, followed by survey deck quality standardization and strengthened positioning as a trusted sur-vey service. The findings indicate that the central strategic challenge for Kospace is not only market visibility, but also the operational reliability of field verification as the foundation of trust-based student accommodation management.
Authors - Sri Bramantoro Abdinagoro, Enda Panggati Abstract - This study examines how audiences process sustainability-oriented hype sneakers through emotional, rational, and hybrid responses in YouTube discourse on the Nike Space Hippie Sneaker. Using the Elaboration Likelihood Model and a semantic NLP approach, this study applies Sentence-BERT (SBERT)-based semantic similarity to identify dual-process consumer responses beyond conventional positive-negative sentiment classification. A corpus of YouTube comments was analyzed using a prototype-based se-mantic embedding approach. Audience comments were classified into emotion-al, rational, hybrid, and ambiguous processing orientations. Robustness checks were conducted using all-MiniLM-L6-v2 and all-mpnet-base-v2 models. The findings show that emotional processing became the most dominant category, followed by hybrid processing, while rational processing appeared in smaller proportions. The results indicate that sustainability in Nike Space Hip-pie discourse is mediated not only by environmental evaluation but also by aesthetic appeal, hype culture, and symbolic sneaker identity. In addition, the emergence of hybrid processing suggests that emotional and rational evaluations may coexist simultaneously within sustainability-oriented sneaker dis-course.
Authors - Darma Rika Swaramarinda, Eka Dewi Utari, Alifah Kusumaningrum, Sri Kartikowati, Muh. Darwis, Triesninda Pahlevi, Zsany Zahra Ailliya Abstract - This study aims to determine the effectiveness and acceptance of Augmented Reality (AR) among office administration lecturers in Indonesia by adopting the Technology Acceptance Model (TAM). This study examines the relationship between Perceived Ease of Use (PEOU), Perceived Effectiveness (PU), Attitude Toward Use (ATU), Behavioral Intention to Use (BITU), and Actual Behavior (AB). Data were collected through a survey of lecturers from three state universities in Indonesia and analyzed using the Partial Least Squares Structural Equation Modeling (PLS-SEM) method. The results showed that Perceived Effectiveness had a positive and significant effect on Attitude Toward Used (β = 0.613; p < 0.001), which indicates that lecturers tend to have a positive attitude towards the use of AR when they feel the benefits of the technology in the learning process. In addition, Attitude Toward Use had a positive and significant effect on Behavioral Intention to Use (β = 0.419; p < 0.001), while Behavioral Intention to Use had a positive and significant effect on Actual Behavior (β = 0.405; p = 0.001). The results of the mediation analysis also showed that Attitude Toward Use partially mediated the relationship between Perceived Effectiveness and Behavioral Intention to Use, while Behavioral Intention to Use mediated the relationship between Attitude Toward Use and Actual Behavior. The results of this study provide strategic implementation for universities in strengthening AR based learning innovation to improve the quality and competitiveness in the era of digital transformation.
Authors - Ain Nasthashia Nasrul, Roy Budiharjo Abstract - This study looks at how corporate governance practices affect financial distress in firms in the energy industry registered as companies under the Indonesia Stock Exchange (IDX) listing throughout the years spanning 2020 until 2024. The study applies institutional shareholding, audit committee capacity, diversity of gender among board directors, and company age as independent variables, while the Altman Z″ Score framework is utilized to evaluate financial dis-tress. This research utilised a quantitative methodology with a causal descriptive framework, employing secondary data sourced from annual reports and financial statements. The sample comprised 28 firms chosen by purposive selection, yielding 140 company-year observations. Results obtained from the model comparison stage revealed the suitability of the Common Effects Model as the selected specification in the panel regression estimation. Research outcomes reveal that corporate governance implementation together with company age significantly affects financial hardship. Meanwhile, institutional ownership, audit committee proportion, and gender composition within the board of directors do not show a statistically meaningful influence on financial distress. On the other hand, financial hardship is positively and significantly impacted by firm age, suggesting that, under some circumstances, older businesses are more likely to face financial trouble. This analytical model is capable of accounting for approximately 24.76% of the changes occurring in financial distress conditions. These results imply that the efficacy and calibre of governance processes are more crucial in reducing financial hardship than the mere existence of a governance organisation.
