Dual graph regularized NMF model for social event detection from Flickr data

Yang, Zhenguo; Li, Qing; Wenyin, Liu; Ma, Yun; Cheng, Min

doi:10.1007/s11280-016-0405-1

Cited by 22 publications

(6 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The prediction of popularity is the key task in the realm of social media analytic. The data provided by Flickr offers photo-sharing services with text descriptions, which has drawn many scholars' research interesting, such as event detection [22,23] and popularity prediction. Kim et al proposed a joint photo stream and blog post framework based on support vector machine [10], which improves the performance on exploration and summarization tasks after using both posted texts and photo streams.…”

Section: Popularity-related Prediction Methodsmentioning

confidence: 99%

Curriculum Learning for Wide Multimedia-Based Transformer with Graph Target Detection

Hong

Huang

et al. 2020

Proceedings of the 28th ACM International Conference on Multimedia

View full text Add to dashboard Cite

The social media prediction task is aiming at predicting content popularity which includes social multimedia data such as photos, videos, and news. The task can not only help make better decisions for recommendation, but also reveals the public attention from evolutionary social systems. In this paper, we propose a novel approach named curriculum learning for wide multimedia-based transformer with graph target detection (CL-WMTG). The curriculum learning is designed for the transformer to improve the efficiency of model convergence. The mechanism of wide multimedia-based transformer is to make the model capable of learning cross information from text, pictures and other features(e.g. categories, location). Moreover, the graph target detection part can extract different features in the picture by pretrained model and reconstruct the features with a homogeneous graph network. We achieved third place in the SMP Challenge 2020.

show abstract

Section: Popularity-related Prediction Methodsmentioning

confidence: 99%

Curriculum Learning for Wide Multimedia-Based Transformer with Graph Target Detection

Hong

Huang

et al. 2020

Proceedings of the 28th ACM International Conference on Multimedia

View full text Add to dashboard Cite

show abstract

“…Event discovery [9][10][11] aims to detect and organize the data distributed on the Internet platforms based on the real-world events they depict. The collected dataset supports event detection from multiple data domains, such as social media sites and news media sites.…”

Section: Application Scenarios and Evaluations 41 Event Discoverymentioning

confidence: 99%

“…There are quite a few existing datasets, e.g., Wikipedia dataset [7], PASCAL sentences [6], NUS-WIDE [3], Flickr-30K [13], MS COCO dataset [4], etc. However, the existing datasets are strongly-aligned paired data [9,10,12], i.e., the textual contents are the exact descriptions of their corresponding images. In reality, the multimodal data may be associated with each other by sharing the same labels, yet they are not trying to describe each other exactly, e.g., news articles on a news media like BBC News reporting a real-world event and images shared by social media users on Flickr that are about the same event.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

MMED: A Multi-domain and Multi-modality Event Dataset

Yang¹,

Lin²,

Cheng³

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

In this work, we construct and release a multi-domain and multimodality event dataset (MMED), containing 25,165 textual news articles collected from hundreds of news media sites (e.g., Yahoo News, Google News, CNN News.) and 76,516 image posts shared on Flickr social media, which are annotated according to 412 real-world events. The dataset is collected to explore the problem of organizing heterogeneous data contributed by professionals and amateurs in different data domains, and the problem of transferring event knowledge obtained from one data domain to heterogeneous data domain, thus summarizing the data with different contributors. We hope that the release of the MMED dataset can stimulate innovate research on related challenging problems, such as event discovery, cross-modal (event) retrieval, and visual question answering, etc.

show abstract

“…The non-negativity makes the matrices easier to inspect. Also, non-negativity is meaningful for the data in actual applications[17].…”

mentioning

confidence: 99%

Pattern Recognition Methods Combined with Raman Spectra Applied to Distinguish Serums from Lung Cancer Patients and Healthy People

Li¹,

Jia²,

Zeng³

et al. 2017

JBM

View full text Add to dashboard Cite

Raman spectroscopy was used to distinguish the serums from lung cancer patients and healthy people, through spectral pretreatment method combined with pattern recognition methods including principal component analysis (PCA), partial least squares-discriminant analysis (PLS-DA), uncorrelated linear discriminant analysis (ULDA), etc. Through the comparisons of the results, it can be found that ULDA and LDA combined with multiple scatter correction (MSC) pretreatment method successfully distinguish the patients of lung cancer and healthy people. The method has academic significance and promising clinical application value.

show abstract

Dual graph regularized NMF model for social event detection from Flickr data

Cited by 22 publications

References 36 publications

Curriculum Learning for Wide Multimedia-Based Transformer with Graph Target Detection

Curriculum Learning for Wide Multimedia-Based Transformer with Graph Target Detection

MMED: A Multi-domain and Multi-modality Event Dataset

Pattern Recognition Methods Combined with Raman Spectra Applied to Distinguish Serums from Lung Cancer Patients and Healthy People

Contact Info

Product

Resources

About