社会计算

实验室面向社会行为、社会媒体和社会事件等开展社会计算研究。针对多源多媒体社会计算数据噪声高、价值密度低的问题，以及行为意图隐蔽模糊、媒体报道内容碎片化、事件记录时空尺度不一致等任务挑战，研究噪声鲁棒学习、多媒体融合、推理学习、多尺度建模等理论与方法。提出了多时空尺度疫情预测、背景感知的多媒体谣言检测等代表性算法。实现了更准确的趋势预测、意图分析、方案决策等。在中国移动用户资费满意度预测中得到成功应用，实现了预测精度的稳定提升。

代表性论文

[1] Penghang Yu, Zhiyi Tan, Guanming Lu, and Bing-Kun Bao*, Mind Individual Information! Principal Graph Learning for Multimedia Recommendation, AAAI Conference on Artificial Intelligence (AAAI) 2025 (Oral)

[2] Qile Fan, Penghang Yu, Zhiyi Tan, Bing-Kun Bao, Guanming Lu*, BeFA: A General Behavior-driven Feature Adapter for Multimedia Recommendation. AAAI Conference on Artificial Intelligence (AAAI) 2025 【论文】

[3] Penghang Yu, Zhiyi Tan, Guanming Lu, Bing-Kun Bao, LD4MRec: Simplifying and Powering Diffusion Model for Multimedia Recommendation. (Arxiv 2023) 【论文】

[4] Penghang Yu, Bing-Kun Bao, Zhiyi Tan, Guanming Lu*. Improving Graph Collaborative Filtering with Directional Behavior Enhanced Contrastive Learning. ACM Transactions on Knowledge Discovery from Data (ACM TKDD) 2024 【论文】

[5] Mingjie Qiu, Zhiyi Tan, Bing-Kun Bao*: MSGNN: Multi-scale Spatio-temporal Graph Neural Network for Epidemic Forecasting. Data Mining and Knowledge Discovery (DMKD) 2024 【论文】【Github】

[6] Penghang Yu, Zhiyi Tan, Guanming Lu, Bing-Kun Bao*. Multi-View Graph Convolutional Network for Multimedia Recommendation. ACM International Conference on Multimedia (ACM MM) 2023 【论文】【Github】

[7] 邱鸣杰, 谭智一, 鲍秉坤*. 城市区域功能感知的细粒度疫情风险评估模型.《中国科技论文》2023 【论文】【Github】

[8] 鲍昕，谭智一*，鲍秉坤，徐常胜. 基于时空注意机制的新冠肺炎疫情预测模型. 《北京航空航天大学学报》2023 【论文】

[9] Junyi Wang, Bing-Kun Bao*, Changsheng Xu. Sentiment-aware Multi-modal Recommendation on Tourist Attractions. MultiMedia Modeling (MMM) 2019 【论文】