A nonparametric mixture model for topic modeling over time

Dubey, Avinava; Hefny, Ahmed; Williamson, Sinead A.; Xing, Eric P.

doi:10.1137/1.9781611972832.59

Cited by 39 publications

(36 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, for each word of each document, a topic is drawn from the mixture and a term is subsequently drawn from the multinomial distribution corresponding to that topic. This has led to the recent development of incorporating temporal dynamics into topic models (e.g., Wang et al 2012;Masada et al 2009;Wang and McCallum 2006;Dubey et al 2013). These models enable us to gain insight into datasets with temporal changes in a convenient way and open future directions for utilizing these models in a more general fashion.…”

Section: Characterizing Temporal Dynamics and Social Responsementioning

confidence: 99%

Topic detection and tracking on heterogeneous information

et al. 2017

View full text Add to dashboard Cite

Given the proliferation of social media and the abundance of news feeds, a substantial amount of real-time content is distributed through disparate sources, which makes it increasingly difficult to glean and distill useful information. Although combining heterogeneous sources for topic detection has gained attention from several research communities, most of them fail to consider the interaction among different sources and their intertwined temporal dynamics. To address this concern, we studied the dynamics of topics from heterogeneous sources by exploiting both their individual properties (including temporal features) and their inter-relationships. We first implemented a heterogeneous topic model that enables topic-topic correspondence between the sources by iteratively updating its topic-word distribution. To capture temporal dynamics, the topics are then correlated with a time-dependent function that can characterise its social response and popularity over time. We extensively evaluate the proposed approach and compare to the state-of-the-art techniques on heterogeneous collection. Experimental results demonstrate that our approach can significantly outperform the existing ones.

show abstract

Section: Characterizing Temporal Dynamics and Social Responsementioning

confidence: 99%

Topic detection and tracking on heterogeneous information

et al. 2017

View full text Add to dashboard Cite

show abstract

“…The second method is modelling the time jointly with the data as in [5], [18], [17]. They treat the timestamp of observations as random variables and infer a distribution over time for each topic.…”

Section: Previous Related Workmentioning

confidence: 99%

“…Since it is an extension of LDA, it is parametric and because of Beta priors, it limits the type of behaviours that the popularity of a topic can show. Dubey et al [5] extended TOT by placing non-parametric priors on topics and time stamps, allowing infinite mixtures of topics for documents, and infinite mixtures of time distributions for topics. Although both limitations are addressed in this model, the way they treat the data is not natural.…”

Section: Previous Related Workmentioning

confidence: 99%

“…θ ji selects a table and enjoys a dish. θ ji selects table b following a CRP as in Equation 5. If b is occupied by other customers, θ ji shares their dish ψ jb otherwise, a new dish is drawn from the global menu following another CRP as in Equation 6.…”

Section: Hierarchical Dirichlet Process Mixture Modelmentioning

confidence: 99%

“…As the result, we relax the constraints that models such as [18], [5] impose on the model. We assume z tji indicates the mixture component to which x tji belongs and definep (k|t) as the empirical probability of topic k in an arbitrary group in time t:…”

Section: F Popularity Of Topicsmentioning

confidence: 99%

See 2 more Smart Citations

Analysing the History of Autism Spectrum Disorder Using Topic Models

Beykikhoshk

Phung

Arandjelović

et al. 2016

2016 IEEE International Conference on Data Science and Advanced Analytics (DSAA)

View full text Add to dashboard Cite

Abstract-We describe a novel framework for the discovery of underlying topics of a longitudinal collection of scholarly data, and the tracking of their lifetime and popularity over time. Unlike the social media or news data, as the topic nuances in science result in new scientific directions to emerge, a new approach to model the longitudinal literature data is using topics which remain identifiable over the course of time. Current studies either disregard the time dimension or treat it as an exchangeable covariate when they fix the topics over time or do not share the topics over epochs when they model the time naturally. We address these issues by adopting a non-parametric Bayesian approach. We assume the data is partially exchangeable and divided it into consecutive epochs. Then, by fixing the topics in a recurrent Chinese restaurant franchise, we impose a static topical structure on the corpus such that the they are shared across epochs and the documents within epochs. We demonstrate the effectiveness of the proposed framework on a collection of medical literature related to autism spectrum disorder. We collect a large corpus of publications and carefully examining two important research issues of the domain as case studies. Moreover, we make the results of our experiment and the source code of the model, freely available to aid other researchers by analysing the results or applying the model to their data collections.

show abstract

Discovering hierarchical topic evolution in time‐stamped documents

Song

Huang

et al. 2015

Asso for Info Science & Tech

View full text Add to dashboard Cite

The objective of this paper is to propose a hierarchical topic evolution model (HTEM) that can organize time‐varying topics in a hierarchy and discover their evolutions with multiple timescales. In the proposed HTEM, topics near the root of the hierarchy are more abstract and also evolve in the longer timescales than those near the leaves. To achieve this goal, the distance‐dependent Chinese restaurant process (ddCRP) is extended to a new nested process that is able to simultaneously model the dependencies among data and the relationship between clusters. The HTEM is proposed based on the new process for time‐stamped documents, in which the timestamp is utilized to measure the dependencies among documents. Moreover, an efficient Gibbs sampler is developed for the proposed HTEM. Our experimental results on two popular real‐world data sets verify that the proposed HTEM can capture coherent topics and discover their hierarchical evolutions. It also outperforms the baseline model in terms of likelihood on held‐out data.

show abstract

A nonparametric mixture model for topic modeling over time

Cited by 39 publications

References 5 publications

Topic detection and tracking on heterogeneous information

Topic detection and tracking on heterogeneous information

Analysing the History of Autism Spectrum Disorder Using Topic Models

Discovering hierarchical topic evolution in time‐stamped documents

Contact Info

Product

Resources

About