Bayesian nonparametric inference of latent topic hierarchies for multimodal data

Over the past few decades, the world has witnessed an unprecedented explosion of information. Of these, a substantial portion consists of unlabeled textual data, such as tweets, news articles, product reviews, and web snippets.
\nAs labeling is extremely expensive, time-consuming, and sometimes biased,
\nhow to effectively analyze these data becomes an imperative.
\nOwing to this, Neural Topic Models (NTMs) have emerged as a promising solution that attracts considerable research attention for their capability and interpretability.
\nThey automatically discover latent topics from unlabeled textual data through neural networks, enabling unsupervised document understanding.
\nThey have derived various downstream applications, such as content recommendation, trend analysis, and text summarization.
\nCompared to conventional topic models like LDA, NTMs offer structural flexibility and also support gradient back-propagation,
\nwhich avoids complicated model-specific derivations and well handle large-scale data.
\n
\nHowever, despite their promise, existing NTMs generally encounter several critical challenges. On the one hand, NTMs often produce low-quality topics, which are even incomparable to conventional models.
\nThese topics tend to be incoherent or repetitive, significantly diminishing their informativeness.
\nOn the other hand, NTMs struggle with low inference ability, leading to less accurate topic distributions for documents.
\nThis limitation greatly hinders document understanding and undermines the subsequent analysis, reasoning, or decision making processes.
\nDue to these challenges, existing NTMs are less useful and applicable for downstream tasks or applications.
\nAs a result, it is necessary to enhance NTMs to deliver more reliable and informative topic modeling.
\n
\nThis thesis aims to advance neural topic modeling by addressing these key challenges.
\nIn particular, we focus on four most popular scenarios: short-texts, cross-lingual, hierarchical, and basic neural topic modeling.
\nFirst, we propose a novel neural topic model tailored for short texts. 
\nThis model leverages a topic-semantic contrastive learning method to captures the similarity relations among short text samples, which works regardless of data augmentation availability.
\nThis refines short text representations, enriches learning signals, thus effectively alleviating the data sparsity issue of short texts and producing informative topics.
\n
\nSecond, we explore dynamic topic modeling.
\nWe present a neural dynamic topic model to track the evolution of dynamic topics by building the contrastive relations among them, rather than relying on the conventional Markov Chains. This model further explicitly excludes unassociated words from dynamic topics to enhance the alignment to their respective time slices.
\nThese improvements enable our model to reliably track topic evolution with high diversity.
\n
\nThird, we focus on the the affinity, rationality, and diversity of hierarchical topic modeling.
\nOur proposed neural hierarchical topic model ensures the sparsity and balance of cross-level topic dependencies using a transport plan dependency method.
\nMoreover, it distributes different semantic granularity to topics at different levels by disentangled decoding.
\nWith these, our model produces affinitive, rational, and diverse topic hierarchies.
\n
\nFourth, we tackle the topic collapsing issue and propose a basic neural topic model.
\nApart from the common reconstruction error, this model introduces a new embedding clustering regularization to force each topic embedding to be the center of a separately aggregated word embedding cluster in the semantic space.
\nThis produces topics with distinct semantics and effectively resolves the topic collapsing issue.
\n
\nFinally, we develop a comprehensive topic modeling toolkit that includes both previous conventional and cutting-edge neural topic models.
\nThis toolkit covers the complete pipelines of topic modeling, such as dataset preprocessing, model training, and evaluation.
\nThese improvements position our toolkit as a valuable resource to accelerate the research and applications of topic models.
\n
\nIn conclusion, this thesis makes significant contributions towards advancing neural topic modeling through multiple models, various scenarios, and a rigorous and comprehensive benchmark toolkit.
\nThese contributions pave the way for the utilization of neural topic modeling in diverse real-world applications.

Topic modeling approaches, such as Latent Dirichlet Allocation (LDA) and Hierarchical LDA (hLDA) have been used extensively to discover topics in various corpora. Unfortunately, these approaches do not perform well when applied to collections of social media posts. Further, these approaches do not allow users to focus topic discovery around subjectively interesting concepts. We propose the new Semi-Supervised Microblog-hLDA (SS-Micro-hLDA) model to discover topic hierarchies in short, noisy microblog documents in a way that allows users to focus topic discovery around interesting areas. We test SS-Micro-hLDA using a large, public collection of Twitter messages and Reddit social blogging site and show that our model outperforms hLDA, Constrained-hLDA, Recursive-rCRP and TSSB in terms of Pointwise Mutual Information (PMI) Score. Further, we test our model in terms of information entropy of held-out data and show that the new approach produces highly focused topic hierarchies.

Local and global topics in text modeling of web pages nested in web sites

A hierarchical latent topic model based on sparse coding

Topic models are often evaluated with measures such as perplexity and topic coherence.However, these methods fall short in determining the comprehensiveness of identified topics.This research introduces a complementary approach to evaluating unsupervised topic models using a labeled dataset.By training hierarchical topic models and utilizing known labels for evaluation, we found a high accuracy of 70% for expected topics.Despite having 90 labels in the dataset, even those representing only 1% of the data achieved an average accuracy of 37.9%, illustrating hierarchical topic models' effectiveness on smaller subsets.Additionally, we confirmed that this new evaluation method helps assess the topic tree quality, demonstrating that hierarchical topic models generate coherent taxonomies.Lastly, we established that coherence measures alone are insufficient for a holistic topic model evaluation.

Disambiguating entity references by annotating them with unique ids from a catalog is a critical step in the enrichment of unstructured content. In this paper, we show that topic models, such as Latent Dirichlet Allocation (LDA) and its hierarchical variants, form a natural class of models for learning accurate entity disambiguation models from crowd-sourced knowledge bases such as Wikipedia. Our main contribution is a semi-supervised hierarchical model called Wikipedia-based Pachinko Allocation Model} (WPAM) that exploits: (1) All words in the Wikipedia corpus to learn word-entity associations (unlike existing approaches that only use words in a small fixed window around annotated entity references in Wikipedia pages), (2) Wikipedia annotations to appropriately bias the assignment of entity labels to annotated (and co-occurring unannotated) words during model learning, and (3) Wikipedia's category hierarchy to capture co-occurrence patterns among entities. We also propose a scheme for pruning spurious nodes from Wikipedia's crowd-sourced category hierarchy. In our experiments with multiple real-life datasets, we show that WPAM outperforms state-of-the-art baselines by as much as 16% in terms of disambiguation accuracy.

Hierarchical topic models, which can extract semantically meaningful topics from a textcorpus in an unsupervised manner and automatically organise them into a topic hierarchy, have been widely used to discover the underlying semantic structure of documents. However, the existing models often assume in the prior that the topic hierarchy is a tree structure, ignoring symmetrical dependenciesbetween topics at the same level. Moreover, the sparsity of text data often complicate the analysis. To address these issues, we propose NSEM-GMHTM as a deep topic model, witha Gaussian mixture prior distribution to improve the model's ability to adapt to sparse data, which explicitly models hierarchical and symmetric relations between topics through the dependency matrices and nonlinear structural equations. Experiments on widely used datasets show that our NSEM-GMHTM generates more coherent topics and a more rational topic structure when compared to state-of-theart baselines. Our code is available at https: //github.com/nbnbhwyy/NSEM-GMHTM.

Topic models are widely used to discover the latent representation of a set of documents. The two canonical models are latent Dirichlet allocation, and Gaussian latent Dirichlet allocation, where the former uses multinomial distributions over words, and the latter uses multivariate Gaussian distributions over pre-trained word embedding vectors as the latent topic representations, respectively. Compared with latent Dirichlet allocation, Gaussian latent Dirichlet allocation is limited in the sense that it does not capture the polysemy of a word such as "bank." In this paper, we show that Gaussian latent Dirichlet allocation could recover the ability to capture polysemy by introducing a hierarchical structure in the set of topics that the model can use to represent a given document. Our Gaussian hierarchical latent Dirichlet allocation significantly improves polysemy detection compared with Gaussian-based models and provides more parsimonious topic representations compared with hierarchical latent Dirichlet allocation. Our extensive quantitative experiments show that our model also achieves better topic coherence and held-out document predictive accuracy over a wide range of corpus and word embedding vectors which significantly improves the capture of polysemy compared with GLDA and CGTM. Our model learns the underlying topic distribution and hierarchical structure among topics simultaneously, which can be further used to understand the correlation among topics. Moreover, the added flexibility of our model does not necessarily increase the time complexity compared with GLDA and CGTM, which makes our model a good competitor to GLDA.

We develop stochastic variational inference, a scalable algorithm for approximating posterior distributions. We develop this technique for a large class of probabilistic models and we demonstrate it with two probabilistic topic models, latent Dirichlet allocation and the hierarchical Dirichlet process topic model. Using stochastic variational inference, we analyze several large collections of documents: 300K articles from Nature, 1.8M articles from The New York Times, and 3.8M articles from Wikipedia. Stochastic inference can easily handle data sets of this size and outperforms traditional variational inference, which can only handle a smaller subset. (We also show that the Bayesian nonparametric topic model outperforms its parametric counterpart.) Stochastic variational inference lets us apply complex Bayesian models to massive data sets.

Semantic multi-grain mixture topic model for text analysis

LDA (Latent Dirichlet Allocation) topic model has been applied into many applications in recent years. But LDA has a shortcoming that it cannot deal with various changes of data set well, which has become a limitation for its applications. Hierarchical Latent Dirichlet Allocation (hLDA) is a generalization of LDA and it can adapt itself to the growing data set automatically. hLDA can mine latent topics from a large amount of discrete data and organize these topics into a hierarchy, in which the topics of higher level are more abstractive while the topics of lower level are more specific. This hierarchy could achieve a deeper semantic model which is similar with human mind. Given a set of documents, hLDA generates a prior distribution of Bayesian nonparametrics using a nested Chinese restaurant process (nCRP)[1]. The documents sharing similar topics are organized into a cluster of path. hLDA learns the distribution of topics using a method of Bayesian posterior inference. This paper tries to study hLDA model in details and apply it into the application of Chinese text clustering. Experiments have shown that hLDA is a very promising model for text clustering.

Digital text is increasing rapidly on the Internet with the excessive use of social media. For this reason, it is very challenging to extract effective information from the digital text due its high dimensionality, sparseness and big data. In this paper, we study the powerful nonparametric Bayesian topic model which is Hierarchical Latent Dirichlet Allocation (hLDA). We deal the issue of learning topics hierarchies from Urdu text data. The presented Topic Model for Urdu is combined with preprocessing activities, hLDA model, and Gibbs Sampling (GS) algorithm. We present hLDA base topic model called Urdu Hierarchical Latent Dirichlet Allocation (uhLDA). Empirical study showed that uhLDA effectively learns the topics hierarchies from 5000 Urdu text documents. Furthermore, we evaluated the results using Pointwise Mutual information (PMI) and it shows that uhLDA outperforms as compared to existing standard topic model LDA.

To derive latent topics from free-text responses on the negative impact of the pandemic on research activities and determine similarities and differences in the resulting themes between academic-based and clinical-based researchers. We performed a secondary analysis of free-text responses from a cross-sectional online survey conducted by the Japan Academy of Nursing Science of its members in early 2020. The participants were categorized into two groups by workplace (academic-based and clinical-based researchers). Latent Dirichlet allocation (LDA) topic modeling was used to extract latent topics statistically and list important keywords/text associated with the topics. After organizing similar topics by principal component analysis (PCA), we finally derived topic-associated themes by reading the keywords/texts and determining the similarity and differences of the themes between the two groups. A total of 201 respondents (163 academic-based and 38 clinical-based researchers) provided free-text responses. LDA identified eight and three latent topics for the academic-based and clinical-based researchers, respectively. While PCA re-grouped the eight topics derived from the former group into four themes, no merging of the topics from the latter group was performed resulting in three themes. The only theme common to the two groups was "barriers to conducting research," with the remaining themes differing between the groups. Using LDA topic modeling with PCA, we identified similarities and differences in the themes described in free-text responses about the negative impact of the pandemic between academic-based and clinical-based researchers. Measures to mitigate the negative impact of pandemics on nursing research may need to be tailored separately.

A massive of text that is generated every minute is increasing dramatically. Therefore, it is more and more important to find an effective model to automatic classify the amount of text. Topic models is the most powerful techniques in text classification. There are many research results in the field of topic model have been published in scholarly journals. The Latent Dirichlet Allocation (LDA) is one of the most popular topic models in text classification. Researchers have proposed many topic evolution models based on LDA to solve some specific problems in applications of text classification. And some joint models which based on topic models combined other algorithms have been studied to enhance the performance of text classification. In this paper, we investigated three categories topic models for text classification and briefly introduced their advantages and disadvantages in the applications of text mining. Also, we introduce the generated process of documents and illustrate the graphical model for each topic models.

This paper deals with the statistical modeling of latent topic hierarchies in text corpora. The height of the topic tree is assumed as fixed, while the number of topics on each level as unknown a priori and to be inferred from data. Taking a nonparametric Bayesian approach to this problem, we propose a new probabilistic generative model based on the nested hierarchical Dirichlet process (nHDP) and present a Markov chain Monte Carlo sampling algorithm for the inference of the topic tree structure as well as the word distribution of each topic and topic distribution of each document. Our theoretical analysis and experiment results show that this model can produce a more compact hierarchical topic structure and captures more fine-grained topic relationships compared to the hierarchical latent Dirichlet allocation model.