A probabilistic approach to spatiotemporal theme pattern mining on weblogs

By Qiaozhu Mei, Chao Liu, Hang Su, Chengxiang Zhai. The Web Conference, 2006

353 citations recorded by the source corpus.

Mining subtopics from weblogs and analyzing their spatiotemporal patterns have applications in multiple domains. In this paper, we define the novel problem of mining spatiotemporal theme patterns from weblogs and propose a novel probabilistic approach to model the subtopic themes and spatiotemporal theme patterns simultaneously. The proposed model discovers spatiotemporal theme patterns by (1) extracting common themes from weblogs; (2) generating theme life cycles for each given location; and (3) generating theme snapshots for each given time period. Evolution of patterns can be discovered by comparative analysis of theme life cycles and theme snapshots. Experiments on three different data sets show that the proposed approach can discover interesting spatiotemporal theme patterns effectively. The proposed probabilistic model is general and can be used for spatiotemporal text mining on any domain with time and location information.

Subfields: Artificial Intelligence, Information Systems, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/a203c6e2933db9c59acd791a8e06f8e9b0270ac2

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 Information Systems

In Statistical and Nonlinear Physics

Citations in the corpus

Topic modeling with network regularization (2008, methodology): “Following [19], we collect weblog articles about a focused topic, by submitting a focused query to Google Blog Search(5), and crawling the content and geographic information of returned blog posts from their original websites.”
Patterns of temporal variation in online media (2011, background): “Research investigated temporal patterns of activity of news articles [5, 30], blogposts [3, 14, 21, 27], Videos [10] and online discussion forums [4].”
Automatic labeling of multinomial topic models (2007, methodology): “Indeed, when labeling topic models manually, most people prefer phrases [17, 16, 18, 24].”
Topic sentiment mixture: modeling facets and opinions in weblogs (2007, methodology): “A lot of previous work has shown the effectiveness of mixture of multinomial distributions (mixture language models) in extracting topics (themes, subtopics) from either plain text collections or contextualized collections [9, 1, 16, 15, 17, 12].”
Discovering geographical topics in the twitter stream (2012, methodology): “We summarize some of representative work here: • Mei et al. [12] propose a model based on Probabilistic Latent Semantic Indexing (PLSA) [11].”
Regions, Periods, Activities: Uncovering Urban Dynamics via Cross-Modal Representation Learning (2017, background): “Geographical topic discovery [31, 19, 35, 15, 36, 24, 34] aims at modeling the topics in different regions.”
Earthquake shakes Twitter users: real-time event detection by social sensors (2010, background): “targeted blogs and analyzed their spatiotemporal patterns [17].”
Placing flickr photos on a map (2009, methodology): “Working with blog data, Mei et al. [12] present methods for ﬁnding latent semantic topics and their distribution over locations (states or countries) and Wang et al. [25] propose a Location-Aware Topic Model based on Latent Dirichlet Al-location.”
A Latent Variable Model for Geographic Lexical Variation (2010, background): “More relevant is the work of Mei et al. (2006), in which the distribution over latent topics in blog posts is conditioned on the geographical location of the author.”
Tweet Analysis for Real-Time Event Detection and Earthquake Reporting System Development (2013, result): “Similarly to our work, Mei et al. targeted blogs and analyzed their spatiotemporal patterns [28].”
TEDAS: A Twitter-based Event Detection and Analysis System (2012, background): “(3) At the system level, we implement a realtime system that can detect and analyze CDEs based on millions of tweets and users.”
The Source-Target Domain Mismatch Problem in Machine Translation (2019, background): “me language, a subject further explored by Britain (2013). Some of these works were the basis for later studies that introduced computational models for how language changes with geographic location (Mei et al., 2006; Eisenstein et al., 2010). Moving to cross-lingual analyses, there has been work at”
A location-based news article recommendation with explicit localized semantic analysis (2013, background): “Most topic models proposed for extracting geographical topics from a location identify topics actually from a set of geo-tagged documents associated with the location [13].”
SAR: A sentiment-aspect-region model for user preference analysis in geo-tagged reviews (2015, background): “Some of these studies [14], [3], [17], [16], [25] focus on analyzing the relation between locations and words/topics without considering users.”
Applications of Topic Models (2017, background): “Internal Labeling Mei et al. [2007b] propose an internal labeling”