Joint Local and Global Sequence Modeling in Temporal Correlation Networks for Trending Topic Detection

Trending topics represent the topics that are becoming increasingly popular and attract a sudden spike in human attention. Trending topics are critical and useful in modern search engines, which can not only enhance user engagements but also improve user search experiences. Large volumes of user search queries over time are indicative aggregated user interests and thus provide rich information for detecting trending topics. The topics derived from query logs can be naturally treated as a temporal correlation network, suggesting both local and global trending signals. The local signals represent the trending/non-trending information within each frequency sequence, and the global correlation signals denote the relationships across frequency sequences. We hypothesize that integrating local and global signals can benefit trending topic detection. In an attempt to jointly exploit the complementary information of local and global signals in temporal correlation networks, we propose a novel framework, Local-Global Ranking (LGRank), to both capture local temporal sequence representation with adversarial learning and model global sequence correlations simultaneously for trending topic detection. The experimental results on real-world datasets from a commercial search engine demonstrate the effectiveness of LGRank on detecting trending topics.

[1]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[2]  Wan Li,et al.  Deal or deceit: detecting cheating in distribution channels , 2014, CIKM.

[3]  Jun Ma,et al.  Time-based Query Classification and its Application for Page Rank , 2011 .

[4]  Zhaochun Ren,et al.  Preserving Local and Global Information for Network Embedding , 2017, ArXiv.

[5]  Yoshua Bengio,et al.  Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling , 2014, ArXiv.

[6]  Gilad Kutiel,et al.  Effective Trend Detection within a Dynamic Search Context , 2016, SIGIR.

[7]  Zhoujun Li,et al.  Emerging topic detection for organizations from microblogs , 2013, SIGIR.

[8]  Jimmy J. Lin,et al.  Mining the Temporal Statistics of Query Terms for Searching Social Media Posts , 2017, ICTIR.

[9]  Jonathan F. Donges,et al.  Comparing linear and nonlinear network construction methods , 2009 .

[10]  Mikhail Belkin,et al.  Laplacian Eigenmaps and Spectral Techniques for Embedding and Clustering , 2001, NIPS.

[11]  G. Caldarelli,et al.  Networks of equities in financial markets , 2004 .

[12]  H. Varian,et al.  Predicting the Present with Google Trends , 2012 .

[13]  Chi-Hoon Lee,et al.  Learning to predict trending queries: classification - based , 2014, WWW '14 Companion.

[14]  Eshel Ben-Jacob,et al.  Dynamics of Stock Market Correlations , 2010 .

[15]  Utkarsh Upadhyay,et al.  Recurrent Marked Temporal Point Processes: Embedding Event History to Vector , 2016, KDD.

[16]  Jeffrey Dean,et al.  Distributed Representations of Words and Phrases and their Compositionality , 2013, NIPS.

[17]  Ke Wang,et al.  TopicSketch: Real-Time Bursty Topic Detection from Twitter , 2013, 2013 IEEE 13th International Conference on Data Mining.

[18]  Yong Tan,et al.  Exploring Correlation Network for Cheating Detection , 2020, ACM Trans. Intell. Syst. Technol..

[19]  Yoshua Bengio,et al.  Professor Forcing: A New Algorithm for Training Recurrent Networks , 2016, NIPS.

[20]  Shuang-Hong Yang,et al.  Mixture of Mutually Exciting Processes for Viral Diffusion , 2013, ICML.

[21]  Eni Mustafaraj,et al.  On the predictability of the U.S. elections through search volume activity , 2011 .

[22]  Le Song,et al.  Wasserstein Learning of Deep Generative Point Process Models , 2017, NIPS.

[23]  Hongyuan Zha,et al.  Modeling the Intensity Function of Point Process Via Recurrent Neural Networks , 2017, AAAI.

[24]  Fernando Diaz,et al.  Time is of the essence: improving recency ranking using Twitter data , 2010, WWW '10.

[25]  R. Mantegna.  Hierarchical structure in financial markets , 1998, cond-mat/9802256.

[26]  Ricardo A. Baeza-Yates,et al.  Query Recommendation Using Query Logs in Search Engines , 2004, EDBT Workshops.

[27]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[28]  Aristides Gionis,et al.  Query-log mining for detecting spam , 2008, AIRWeb '08.

[29]  Ziad Al Bawab,et al.  Finding trending local topics in search queries for personalization of a recommendation system , 2012, KDD.

[30]  Gilad Mishne,et al.  Towards recency ranking in web search , 2010, WSDM '10.

[31]  Doug Beeferman,et al.  Agglomerative clustering of a search engine query log , 2000, KDD '00.

[32]  Andrew C. Miller,et al.  Advances in nowcasting influenza-like illness rates using search query logs , 2015, Scientific Reports.

[33]  Hongbo Deng,et al.  Learning Parametric Models for Context-Aware Query Auto-Completion via Hawkes Processes , 2017, WSDM.

[34]  Shubhra Kanti Karmaker Santu,et al.  Modeling the Influence of Popular Trending Events on User Search Behavior , 2017, WWW.

[35]  Yehuda Koren,et al.  Expediting search trend detection via prediction of query counts , 2013, WSDM.

[36]  Fabrizio Lillo,et al.  Correlation, Hierarchies, and Networks in Financial Markets , 2008, 0809.4615.

[37]  Shlomo Havlin,et al.  Very early warning of next El Niño , 2014, Proceedings of the National Academy of Sciences.

[38]  H. Varian,et al.  Predicting the Present with Google Trends , 2009 .

[39]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[40]  Tao Mei,et al.  Learning to personalize trending image search suggestion , 2014, SIGIR.

[41]  Le Song,et al.  Learning Conditional Generative Models for Temporal Point Processes , 2018, AAAI.