5th World Conference on Information Systems for Business Management: Full Schedule

2:58pm PDT

Opening Remarks

Monday October 19, 2026 2:58pm - 3:00pm PDT

Virtual Room F

Monday October 19, 2026 2:58pm - 3:00pm PDT
Virtual Room F Bangkok, Thailand

Virtual Room 9F, Virtual Room F

3:00pm PDT

Efficient Car Logo Detection via YOLOv8 and Attention Mechanism Fusion

Monday October 19, 2026 3:00pm - 5:00pm PDT

Virtual Room F

Authors - Nhat Minh Nguyen, Tu Anh Nguyen, Huy Thai Hinh, Quang Nhat Nguyen, Thong Minh Phuc Nguyen, Vinh Dinh Nguyen
Abstract - Car logo detection is crucial for intelligent transportation systems. This paper compares attention mechanisms integrated into YOLOv8m for car logo detection, evaluating Efficient Channel Attention (ECA), SimAM, and their combination against a baseline YOLOv8m (95.2% [email protected], 48.6% [email protected]:0.95, 79.3 GFLOPs). On a dataset of 5372 images, ECA achieves 94.5% [email protected] and 50.2% [email protected]:0.95 with 79.1 GFLOPs, SimAM reaches 96.1% [email protected] and 50.0% [email protected]:0.95 with 79.1 GFLOPs, and the combined approach attains 95.3% [email protected] and 50.1% [email protected]:0.95 with 79.3 GFLOPs. All variants maintain real-time performance with inference speeds of 29.29 FPS (ECA), 28.69 FPS (SimAM), and 27.95 FPS (combined), compared to the baseline’s 28.32 FPS on an NVIDIA RTX 3050 laptop GPU. These results demonstrate that integrating lightweight attention mechanisms can substantially enhance detection performance in data-constrained scenarios while preserving real-time efficiency.

Paper Presenter

Nhat Minh Nguyen

Vietnam

Monday October 19, 2026 3:00pm - 5:00pm PDT
Virtual Room F Bangkok, Thailand

Virtual Room 9F, Virtual Room F

3:00pm PDT

Empowering Auditor Performance Through Digital Access and IT Infrastructure: The Mediating Role of Auditor Competence

Monday October 19, 2026 3:00pm - 5:00pm PDT

Virtual Room F

Authors - Windy Permata Suyono, Dwi Handarini, Eka Septariana Puspa, Surya Anugrah, Wida Aristanti, Rio Firnanda
Abstract - This study investigates the influence of digital access and IT infra-structure on auditor performance, with auditor competence as a mediating variable. Grounded in the Technology–Organization–Environment (TOE) and Task–Technology Fit (TTF) frameworks, data were collected from 110 auditors in Jakarta and West Java using a structured online survey. Structural Equation Modelling with Partial Least Squares (SEM-PLS) was employed to analyse the relationships. The results reveal that while digital access does not have a direct effect on auditor performance, it significantly influences auditor competence, which in turn enhances performance. IT infrastructure shows both direct and indirect effects through auditor competence. These findings underscore the critical role of auditor competence in leveraging technological resources to achieve optimal performance. The study provides practical insights for audit institutions aiming to enhance audit effectiveness through digital capability development and infra-structure investment. It also contributes to the growing literature on digital trans-formation in the auditing profession, especially within the context of emerging economies.

Paper Presenter

Windy Permata Suyono

Indonesia

Monday October 19, 2026 3:00pm - 5:00pm PDT
Virtual Room F Bangkok, Thailand

Virtual Room 9F, Virtual Room F

3:00pm PDT

Fine-Tuning BART for Multi-Label ICD-9 Prediction from Clinical Discharge Summaries

Monday October 19, 2026 3:00pm - 5:00pm PDT

Virtual Room F

Authors - Ch Geethika Gayatri, P Lavanya, S Manvitha Reddy, Nikitha K
Abstract - Accurate and automated assignment of ICD-9 codes from clinical narratives is essential for healthcare analytics, and clinical decision support. However, the task remains complex due to the high dimensionality of the label space, variability in clinical language, and the multi-label nature of diagnostic documentation. This paper presents a novel deep learning framework for multi-label ICD-9 code prediction from discharge summaries using a fine-tuned Bidirectional and Auto-Regressive Transformers(BART). Targeting real-world clinical documentation scenarios, the proposed approach models unstructured medical narratives using a sequence-to-sequence architecture that captures both global context and fine-grained semantic cues. Each discharge summary is tokenized and encoded using BART’s bidirectional encoder and autoregressive decoder, enabling robust multi-label classification across the top 20 most frequent ICD-9 codes. Evaluation on the MIMIC-III dataset demonstrates strong performance, achieving a F1 score of 71%, while also showing balanced precision-recall behavior across high and low-frequency codes. ROC-AUC confirms the model’s discriminative capability across imbalanced labels. These results validate the model’s capability to handle complex, multi-label classification scenarios common in clinical documentation.

Paper Presenter

S Manvitha Reddy

India

Monday October 19, 2026 3:00pm - 5:00pm PDT
Virtual Room F Bangkok, Thailand

Virtual Room 9F, Virtual Room F

3:00pm PDT

Formative Assessment of Spoken English Using Large Language Models in a Controlled Intervention Study

Monday October 19, 2026 3:00pm - 5:00pm PDT

Virtual Room F

Authors - Mahimaben Panjabi, Vijay Makwana
Abstract - This study investigated the efficacy of Large Language Models (LLMs) for providing formative assessment of spoken English within a mixed-methods, one-group pre-test/post-test design. The research involved 30 intermediate English as a Second Language (ESL) learners and five expert faculty members. The design integrated quantitative pre-test/post-test data with qualitative analysis of AI-generated feedback, student reflections, and faculty interviews. During the intervention, all students used LLM tools (ChatGPT and Google Gemini) for practice and feedback. To ensure a robust, triangulated assessment, a coder-based framework was implemented where human experts evaluated spoken tasks against linguistic criteria. A paired-samples t-test revealed a statistically significant improvement in speaking proficiency from pre-test to post-test. Thematic analysis of qualitative data indicated that LLM-generated feedback was perceived as useful, accessible, and effective in reducing learner anxiety. This multi-faceted approach affirms that LLMs can be effective supplementary tools in language education and offers a holistic model for their integration into formative assessment practices.

Paper Presenter

Mahimaben Panjabi

India

Monday October 19, 2026 3:00pm - 5:00pm PDT
Virtual Room F Bangkok, Thailand

Virtual Room 9F, Virtual Room F

3:00pm PDT

Nonlinear Classification on MiniRocket Architecture for Fish-Freshness Prognostics

Monday October 19, 2026 3:00pm - 5:00pm PDT

Virtual Room F

Authors - Raghavan Vaidhyaraman, Aryavardhan Modi, Jayakumar Kaliappan
Abstract - This paper aims to evaluate the performance of multiple non-linear classification techniques. Time-series classification has become central to many sensor-driven applications, where capturing temporal patterns efficiently and accurately is critical. In this paper, we evaluate and extend the proposed MiniRocket framework for classifying multivariate sensor data, focusing on a practical “Fish Freshness” monitoring use case. Precise quantification of post-harvest fish freshness underpins foodsafety compliance and public-health protection while curbing spoilageinduced economic losses across the cold-chain. In operational practice, high-fidelity freshness classification enables dynamic inventory routing, just-in-time processing, and trustworthy quality labelling, thereby reducing food waste and reinforcing consumer confidence throughout seafood supply networks. MiniRocket is a highly efficient convolution-based feature extractor that transforms each raw time series into a fixed-length vector of pattern-frequency features (Proportion of Positive Values). We demonstrate that, when paired with a simple linear classifier such as Ridge Classifier, MiniRocket achieves state-of-the-art accuracy while requiring orders of magnitude less computation than deep-learning alternatives. We have demonstrated that combining MiniRocket with Random Forest Classifier yields greater than or equal to 90% classification accuracy, particularly improving performance on ambiguous classes like ”Semi-Fresh.” The model’s simplicity, scalability, and efficiency make it suitable for real-time applications. Our findings validate MiniRocket as a practical solution for robust, low-latency time-series classification in resource-constrained environments.

Paper Presenter

Raghavan Vaidhyaraman

India

Monday October 19, 2026 3:00pm - 5:00pm PDT
Virtual Room F Bangkok, Thailand

Virtual Room 9F, Virtual Room F

3:00pm PDT

Optimizing Feature Selection for Medical Diagnosis Systems Using Differential Evolution

Monday October 19, 2026 3:00pm - 5:00pm PDT

Virtual Room F

Authors - GGS Pradeep, Thrilok. Kolla, N Vijayalakshmi, Rajesh Sharma R
Abstract - Machine learning has aided the improvement of medical diagnostic systems that help to detect diseases accurately in the present era. The work gave a mathematical and algorithmic handbook for characteristic selection with a binary Genetic Algorithm (GA) of particular use to jobs involving medical diagnosis. Clinical datasets usually have many dimensions and redundancy, which most time affects model performance and increases computational complexity. In this study is presented a mathematical and algorithmic guide for selecting features through a binary Genetic Algorithm (GA), especially suitable for medical diagnostic tasks. The proposed method seeks to obtain the most informative subset of features by optimizing a fitness function that balances between classification accuracy and dimensionality reduction. The work develops a thorough mathematical model that incorporates data preprocessing, binary encoding of feature subsets, and repetitive evolutionary optimization. The metric used in testing the classification model is standard performance metrics-accuracy, sensitivity, specificity, and F1 score, summarized using a confusion matrix. Understanding feature selection stability across generations is also examined and visualized in a multidimensional performance space. The results would indicate a direct convergence of the model into high-performing feature configurations while avoiding risk for overfitting. An interpretative paradigm, speed-up in computations, as well as high reliability in diagnosis, therefore pushing this method as a tool of great value in clinical decision support systems.

Paper Presenter

GGS Pradeep

India

Monday October 19, 2026 3:00pm - 5:00pm PDT
Virtual Room F Bangkok, Thailand

Virtual Room 9F, Virtual Room F

3:00pm PDT

Optimizing Multi-Agent and Functional Architectures for Enhanced Financial Trading Performance: VinTradeAgent Case Study

Monday October 19, 2026 3:00pm - 5:00pm PDT

Virtual Room F

Authors - An Dinh Van, Anh Nguyen Thi Linh, Phuong Pham Nguyen Hien, Hung Nguyen Gia, Trinh Tran Thi Kieu, Hung Nguyen Quang
Abstract - This study proposes and evaluates a hybrid framework that integrates multi-agent systems and functional architecture for automated financial trading in Vietnam. The framework assigns specialized roles - fundamental, technical, news, market, trader and risk agents - operating over multimodal stock market data (OHLCV, technical indicators, firm fundamentals, and textual news). Agents interact via structured horizontal debates and vertical risk gating to produce auditable trading proposals. Experimental results demonstrate improved risk-adjusted returns and enhanced explainability, while modular design supports deployment for back-office automation (e.g., report generation and opportunity identification). The study contributes a methodology for combining agentic large language model (LLMs) with disciplined functional pipelines and provides practical guidance for deploying such systems in emerging markets characterized by high volatility and limited liquidity. Implications for FinTech adoption and regulatory alignment in Vietnam are discussed. Vin-Trade-Agent is available at https://github.com/thanhENC/Vin-Trade-Agent.

Paper Presenter

An Dinh Van

Vietnam

Monday October 19, 2026 3:00pm - 5:00pm PDT
Virtual Room F Bangkok, Thailand

Virtual Room 9F, Virtual Room F

3:00pm PDT

Predictive Modeling of Customer Churn and Personalized Subscription Recommendations for OTT Platforms

Monday October 19, 2026 3:00pm - 5:00pm PDT

Virtual Room F

Authors - Kavitha Dhanushkodi, Uma Sankary A
Abstract - With the explosive growth of Over-the-Top (OTT) plat- forms, reducing subscriber churn has emerged as a central problem for service providers. This paper introduces a single framework for churn pre- diction and customized subscription plan recommendation, capitalizing on the advantages of Graph Neural Networks (GNN) and Reinforcement Learning (RL). Churn prediction was carried out on four models: Random Forest, XGBoost, Long Short-Term Memory (LSTM), and GNN, with the GNN model showing better ability to detect potential churners by being able to model rich relationships in user data. For personalized recommendations, both Cosine Similarity-based and Deep Q-Learning (DQN)-based approaches were utilized, with DQN providing dynamic, user-specific plan recommendations. The model was trained and tested on a balanced dataset with 147,269 training samples and 25,000 test samples, and performance was gauged using accuracy, recall, F1-score, and AUC-ROC. Analysis shows that the use of GNN for churn prediction and DQN for recommendation enhances the efficacy of retention mechanisms and increases user satisfaction. The technique empowers OTT providers with an effective means of proactive user interaction and churn prevention.

Paper Presenter

Uma Sankary A

India

Monday October 19, 2026 3:00pm - 5:00pm PDT
Virtual Room F Bangkok, Thailand

Virtual Room 9F, Virtual Room F

3:00pm PDT

Strategic Media Insights on Video-on-Demand (VoD) Platforms through Graph Analytics

Monday October 19, 2026 3:00pm - 5:00pm PDT

Virtual Room F

Authors - Ghanem Ghanem Binhamooda AlDhaheri, Khalifa Mohamed Abdulla Alansi, Hazza Meqbel Ali Alameri, Gurdal Ertek, Ananth Chiravuri
Abstract - In today’s digital entertainment landscape, video-on-demand (VoD) streaming platforms are essential for accessing a broad range of content. This study examines the content of extensive media catalogs available on some of the major streaming platforms, namely Amazon, Apple, Disney, HBO, Netflix, and Paramount. We used a structured dataset and custom-developed a graph analytics methodology that provided a thorough analysis of the data. The sample collection of algorithmically generated graph visualizations help draw basic and in-depth insights, targeted to support informed decision-making for content creators, platform managers, and viewers alike.

Paper Presenter

Ananth Chiravuri

United Arab Emirates

Monday October 19, 2026 3:00pm - 5:00pm PDT
Virtual Room F Bangkok, Thailand

Virtual Room 9F, Virtual Room F

3:00pm PDT

The 'Grammarly Effect': A Mixed-Methods Study on the Impact of Instantaneous AI Feedback on Student Writing Anxiety and Revision Processes

Monday October 19, 2026 3:00pm - 5:00pm PDT

Virtual Room F

Authors - Kinjal Bhatia
Abstract - The integration of AI-powered writing assistants like Grammarly into academic life has fundamentally altered the writing landscape. While these tools provide immediate feedback, their influence on the psychological and procedural aspects of student writing is not well understood. This paper introduces the "Grammarly Effect," examining it through a sequential explanatory mixed-methods study. The research investigated how real-time AI suggestions affect the writing anxiety and revision habits of undergraduate students. In the quantitative phase (N=124), a quasi-experimental de-sign using pre- and post-task surveys found a significant decrease in anxiety associated with surface-level correctness (e.g., grammar, spelling). However, the qualitative phase, involving think-aloud protocols and interviews with a subset of students (n=15), revealed a more nuanced situation. While anxiety over superficial errors di-minished, a reliance on the tool fostered a different anxiety related to authorial originality and self-trust. Analysis of revision behaviors showed a consistent pattern of uncritical acceptance of AI suggestions, prioritizing surface-level fixes over deep, rhetorical engagement. The study concludes that the "Grammarly Effect" is a paradoxical phenomenon, highlighting a critical need for pedagogical approaches that foster AI literacy, empowering students to use these tools as aids rather than oracles.

Paper Presenter