Event Detection in Twitter

By Jianshu Weng, Bu-Sung Lee. International Conference on Web and Social Media, 2011

866 citations recorded by the source corpus.

Twitter, as a form of social media, is fast emerging in recent years. Users are using Twitter to report real-life events. This paper focuses on detecting those events by analyzing the text stream in Twitter. Although event detection has long been a research topic, the characteristics of Twitter make it a non-trivial task. Tweets reporting such events are usually overwhelmed by high flood of meaningless “babbles”. Moreover, event detection algorithm needs to be scalable given the sheer amount of tweets. This paper attempts to tackle these challenges with EDCoW (Event Detection with Clustering of Wavelet-based Signals). EDCoW builds signals for individual words by applying wavelet analysis on the frequencybased raw signals of the words. It then filters away the trivial words by looking at their corresponding signal autocorrelations. The remaining words are then clustered to form events with a modularity-based graph partitioning technique. Experimental results show promising result of EDCoW.

Subfields: Artificial Intelligence, Computer Networks and Communications, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/01ff2b834772dfc2b8b7ba00620b65abb9444a75

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Networks and Communications

In Statistical and Nonlinear Physics

Citations in the corpus

Exploiting Social Relations and Sentiment for Stock Prediction (2014, background): “Related works include detecting and summarizing events (Weng and Lee, 2011; Weng et al., 2011; Baldwin et al., 2012; Gao et al., 2012) and analyzing sentiments about them (Pang and Lee, 2008; Liu, 2012), etc.”
Spatiotemporal social media analytics for abnormal event detection and examination using seasonal-trend decomposition (2012, background): “With Senseplace2, MacEachren et al. [15] demonstrate a visualization system that denotes the message density of actual or textually inferred Twitter message locations.”
Learning similarity functions for topic detection in online reputation monitoring (2014, methodology): “Co-occurrence bursts and temporal signals have been adopted for detecting topics in both the blogosphere [30, 16, 37] and the Twittersphere [25, 8, 37].”
Semantics-driven Event Clustering in Twitter Feeds (2015, methodology): “Such a technique was successfully used by Weng et al. in their EDCoW algorithm to detect Twitter events [21], and by Chen and Roy to detect events in Flickr photo collections on a geographic scale [6].”
Real-Time Detection of Traffic From Twitter Stream Analysis (2015, background): “According to Pear Analytics, 1 it has been estimated that over 40% of all Twitter 2 SUMs (i.e., tweets ) is pointless with no useful information for the audience, as they refer to the personal sphere [16].”
Twevent: segment-based event detection from tweets (2012, methodology): “This dataset was built by Weng and Lee for evaluating the EDCoW event detection method in [21].”
Sensing Trending Topics in Twitter (2013, methodology): “[32] compute (a variant of ) for each term in each considered time slot, and apply wavelet analysis on consecutive blocks.”
Multi-Task Learning for Spatio-Temporal Event Forecasting (2015, methodology): “Clustering-based methods search for novel spatial clusters of documents or terms using predeﬁned similarity metrics, such as cosine similarity and social similarity for documents [3], or auto-correlations [2] and co-occurrences [32] for terms.”
Harvesting ambient geospatial information from social media feeds (2011, background): “…Sankar-anarayanan et al. (2009) presented an approach to extract news content from noisy twitter feeds guided by a small sample of manually identiﬁed seeders which are in essence news-oriented tweeters, and Weng and Lee (2011) presented an approach based on clustering of wavelet-based signals.”
Social media analytics - Challenges in topic discovery, data collection, and data preparation (2018, methodology): “All three of them motivate or discuss their models in the context of each of the challenges, from the ever increasing amount of real-time data to the variable and dynamic nature of the data and the noise in tweets in the form of “ pointless ‘ babbles ” ' (Weng et al., 2011).”
Processing Social Media Messages in Mass Emergency (2014, methodology): “Weng et al. [2011] present an algorithm for event detection from tweets using clustering of wavelet-based signals.”
Fine-grained location extraction from tweets with temporal awareness (2014, result): “This observation is consistent with the earlier finding that Twitter is an individual news media [10, 25, 29].”
Spatio-Temporal Data Mining (2017, background): “…of crime data [Eftelioglu et al. 2014], twitter data [Abdelhaq et al. 2013; Chierichetti et al. 2014; Ihler et al. 2006; Walther and Kaisser 2013; Weng and Lee 2011], geo-tagged photos [Zheng et al. 2012], traffic accidents [Zheng et al. 2012], and epidemiological data [Glatman-Freedman et al.…”