Abstract:A key challenge in mining social media data streams is to identify events which are actively discussed by a group of people in a specific local or global area. Such events are useful for early warning for accident, protest, election or breaking news. However, neither the list of events nor the resolution of both event time and space is fixed or known beforehand. In this work, we propose an online spatio-temporal event detection system using social media that is able to detect events at different time and space resolutions. First, to address the challenge related to the unknown spatial resolution of events, a quad-tree method is exploited in order to split the geographical space into multiscale regions based on the density of social media data. Then, a statistical unsupervised approach is performed that involves Poisson distribution and a smoothing method for highlighting regions with unexpected density of social posts. Further, event duration is precisely estimated by merging events happening in the same region at consecutive time intervals. A post processing stage is introduced to filter out events that are spam, fake or wrong. Finally, we incorporate simple semantics by using social media entities to assess the integrity, and accuracy of detected events. The proposed method is evaluated using different social media datasets: Twitter and Flickr for different cities: Melbourne, London, Paris and New York. To verify the effectiveness of the proposed method, we compare our results with two baseline algorithms based on fixed split of geographical space and clustering method. For performance evaluation, we manually compute recall and precision. We also propose a new quality measure named strength index, which automatically measures how accurate the reported event is.

An SVD-based Multimodal Clustering method for Social Event Detection

Real-Time Multimedia Social Event Detection in Microblog

Multi-Modal Supervised Latent Dirichlet Allocation for Event Classification in Social Media

Social Event Classification Via Boosted Multimodal Supervised Latent Dirichlet Allocation.

Modality Mixture Projections for Semantic Video Event Detection

Boosted Multi-modal Supervised Latent Dirichlet Allocation for Social Event Classification

Columbia-UCF TRECVID2010 Multimedia Event Detection: Combining Multiple Modalities, Contextual Concepts, and Temporal Matching.

Multimodal Classification of Events in Social Media

Bi-Level Semantic Representation Analysis for Multimedia Event Detection

Multiscale event detection in social media

Discriminative Multimodal Embedding for Event Classification

Multimodal Classification of Urban Micro-Events

Semantic event detection via multimodal data mining

Effective video event detection via subspace projection

Multimodal Sparse Coding for Event Detection

Real-time Spatio-temporal Event Detection on Geotagged Social Media

Embed2Detect: temporally clustered embedded words for event detection in social media

Resource Constrained Multimedia Event Detection

Special Issue on Multimedia Event Detection

Multimodal Collaboration Networks for Geospatial Vehicle Detection in Dense, Occluded, and Large-Scale Events

Multi-Modal Event Topic Model for Social Event Analysis.