This three-volume set LNCS 16431-16433 constitutes the refereed proceedings of the 12th International Conference on Behavioural and Social Computing, BESC 2025 held in Hong Kong SAR, China, during October 16–18, 2025.
The 18 full papers and 5 short papers presented in this volumes were carefully reviewed and selected from 149 submissions.
These papers focus on various aspects of social computing, applied psychology, behaviour pattern learning, text and sentiment
analysis, graph learning, AI in education, and AI for social good.
Table of Contents:
.- Early Detection of Fake News Based on Sequential Analysis of Social Context and Adaptive Thresholding.
.- Accelerating High-Dimensional Expensive Multi-Objective Optimization via Surrogate-Assisted Fuzzy Directed Sampling.
.- An Improved Large-Scale Multi-Objective Competitive Swarm Optimizer Based on Harris Hawks Optimization.
.- Multimodal Multiobjective Optimization Algorithm Based on Two-Stage Proportional Selection.
.- A Multidimensional Sentiment Analysis Model for Experience Assessment.
.- Discrete-Time Sigma-Delta Modulator Coefficient Optimization via Soft Actor-Critic Reinforcement Learning.
.- Anomaly Detection Method for Tabular Data Based on a Three-Phase Learning Paradigm.
.- GADM: Granularity-Aware Diffusion Model for Uncertainty Forecasting in Non-Stationary Time Series.
.- SpiderGNN: Spatial-Aware Predictive Inference with Dynamic Edge Reasoning for In-Building 5G Signal Estimation.
.- Contrastive Representation Learning of Floor Plan Images using Human Annotated Attributed Graphs.
.- AgentLesson: LLM-Based Multi-Agent System for Educational Lesson Plan Generation.
.- Bridging Dependency Enhanced Counterfactual Supporting for Multi-hop Question Generation.
.- NLCC: Noisy Label Correction with CLIP for Robust Few-Shot Learning.
.- Quantifying Off-ball Attacking Contribution in Football: A GCN-LSTM Approach Supervised by Expert Assessments.
.- Optimizing Urban Commercial Landscapes with POI Strategic Group Analysis: A Meta-Learning Hypergraph Approach.
.- DCSN: Dynamic Calibration and Contrastive Sharpening for MultiModal Recommendation.
.- Echoes or Adversaries? Asymmetric Polarization in Elite Digital Communication.
.- Student Learning Engagement Recognition Method Based on Video Transformer.
.- Interpretable and Transparent Experimentation on Policy Ideas in a Social Simulation Environment.
.- Do You Tolerate System Failures? An Experimental Study on the Influence of Providing Prior Statements on Human Evaluation of System Failures.
.- Classroom-Free Educational Experimentation: A Stochastic Learning-Design Simulation Framework with LLM Optimization.
.- An Examination of Methods for Resolving Reviewer Dissatisfaction in Hotel Recommendation Systems.
.- Developing a Novel Tag Cloud System for Optimized Visualization and Behavioural Analytics in Cosmetic Review Platforms.
.- Leveraging Comment Data for Enhanced Content Discovery through Time Series Analysis of Impressive Scenes in Videos.
.- Predicting Career Trajectories After Career Breaks: A Data-Driven Analysis Using LinkedIn Profiles.
.- Framework for Diverse Depression Patients Roleplaying and Cognitive Diagnosis of Scales Using LLMs Based on CoT Prompts.
.- Using Topic Detection to Analyze Student Reflections in Service-Learning: A Text Mining Method for Understanding Learning Gains.
.- Modeling Rater Severity Drift in Single-Rater Performance Assessments Using Bayesian Hierarchical Methods.
.- Continuous Sub-sequence Denoising For Sequential Recommendation.
.- ReCareer: Hybrid Graph Neural Networks for Post-Career-Break Job Recommendation.
.- Two-Lane Local Optimality Driving Strategy for Enhancing Traffic Stability.
.- Multimodal Learning Analytics for Predicting Learning Gains in Online Service-Learning Programs.
.- Beyond Words: Latent Profiles of Online Communication Styles and their Association with Social Well-Being on Social Media.
