Multimodal Classification of Events in Social Media
St. Pölten University of Applied Sciences(圣珀尔滕应用科学大学);Vienna University of Technology(维也纳工业大学)
Matthias Zeppelzauer, Daniel Schopfhauser
备注:
Preprint of accepted manuscript for the Elsevier Image and Vision Computing Journal (IMAVIS). The paper will be published by IMAVIS under DOI 10.1016/j.imavis.2015.12.004
A large amount of social media hosted on platforms like Flickr and Instagram is related to social events. The task of social event classification refers to the distinction of event and non-event-related content as well as the classification of event types (e.g. sports events, concerts, etc.). In this paper, we provide an extensive study of textual, visual, as well as multimodal representations for social event classification. We investigate strengths and weaknesses of the modalities and study synergy effects between the modalities. Experimental results obtained with our multimodal representation outperform state-of-the-art methods and provide a new baseline for future research.
A Multimodal Approach to Predict Social Media Popularity
Nirma Institute of Technology(尼玛理工学院);Delhi Technological University(德里理工大学);UALR(阿肯色大学小石城分校);National University of Singapore(新加坡国立大学);IIIT-Delhi(德里英迪拉·甘地信息技术学院)
多个模态表示数据源传递信息的不同方面。现代社交媒体平台是多模态数据的主要来源之一,其中用户通过发布文本以及多媒体内容(例如图像和视频)来使用不同的表达模式以共享信息。这些帖子中包含的多模式信息可能有助于预测其受欢迎程度。据我们所知,不存在这样的多模态数据集来预测社交媒体照片。在这项工作中,我们提出了一个多模态数据集consistiting的内容,上下文和社会信息的流行预测。具体来说,我们在ACM Multimedia Grand Challenge 2017中使用图像内容,标题,描述和标签来增强SMPT 1数据集的社交媒体预测。接下来,在本文中,我们提出了一种利用视觉特征(即,内容信息),文本特征(即,上下文信息),以及社交特征(例如,平均视图和组计数)来预测社交媒体照片在视图计数方面的流行度。实验结果证实,尽管我们的多模态方法使用了来自SMP-T1的训练数据集的一半,但它实现了与最先进的方法相当的性能。
Social-MAE: A Transformer-Based Multimodal Autoencoder for Face and Voice
Numediart Institute, ISIA Lab, University of Mons(Numediart研究院、ISIA实验室、蒙斯大学);Institute for Creative Technologies, University of Southern California(创意技术研究所、南加州大学)
MM-Claims: A Dataset for Multimodal Claim Detection in Social Media
TIB – Leibniz Information Centre for Science and Technology(莱布尼茨科技信息中心(TIB));L3S Research Center, Leibniz University Hannover(汉诺威莱布尼茨大学L3S研究中心);Jozef Stefan Institute(约瑟夫·斯特凡研究所)
Automatic Modeling of Social Concepts Evoked by Art Images as Multimodal Frames
Department of Computer Science and Engineering (DISI), University of Bologna, Italy(博洛尼亚大学计算机科学与工程系(DISI));Department of Modern Languages, Literatures and Cultures (LILEC), University of Bologna, Italy(博洛尼亚大学现代语言、文学与文化系(LILEC))