Lessons from the Bible on Modern Topics: Low-Resource Multilingual
            Topic Model Evaluation

Hao, Shudong; Boyd-Graber, Jordan; Paul, Michael J.

doi:10.18653/v1/n18-1099

Cited by 11 publications

(8 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We choose topic coherence (Hao, Boyd-Graber, and Paul 2018) and crosslingual document classification (Smet, Tang, and Moens 2011) as intrinsic and extrinsic evaluation tasks, respectively. The reason for choosing these two tasks is that they examine the models from different angles: Topic coherence looks at topic-word distributions, whereas classification focuses on document-topic distributions.…”

Section: Discussionmentioning

confidence: 99%

An Empirical Study on Crosslingual Transfer in Probabilistic Topic Models

Hao

Paul²

2020

Computational Linguistics

Self Cite

View full text Add to dashboard Cite

Probabilistic topic modeling is a common first step in crosslingual tasks to enable knowledge transfer and extract multilingual features. While many multilingual topic models have been developed, their assumptions about the training corpus are quite varied, and it is not clear how well the different models can be utilized under various training conditions. In this paper, we systematically study the knowledge transfer mechanisms behind different multilingual topic models, and through a broad set of experiments with four models on ten languages, we provide empirical insights that can inform the selection and future development of multilingual topic models.

show abstract

Section: Discussionmentioning

confidence: 99%

An Empirical Study on Crosslingual Transfer in Probabilistic Topic Models

Hao

Paul²

2020

Computational Linguistics

Self Cite

View full text Add to dashboard Cite

show abstract

“…Additionally, we propose a supervised topic quality estimator that by combining multiple metrics delivers even better results. For future work, we intend to work with larger datasets to investigate neural solutions to combine features from different metrics, as well as to apply our findings to other variants of LDA models trained on low-resource languages, where high-quality external corpora are usually not available (Hao et al, 2018).…”

Section: Discussionmentioning

confidence: 99%

Evaluating Topic Quality with Posterior Variability

Xing¹,

Paul

Carenini

2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

Self Cite

View full text Add to dashboard Cite

Probabilistic topic models such as latent Dirichlet allocation (LDA) are popularly used with Bayesian inference methods such as Gibbs sampling to learn posterior distributions over topic model parameters. We derive a novel measure of LDA topic quality using the variability of the posterior distributions. Compared to several existing baselines for automatic topic evaluation, the proposed metric achieves state-of-the-art correlations with human judgments of topic quality in experiments on three corpora. 1 We additionally demonstrate that topic quality estimation can be further improved using a supervised estimator that combines multiple metrics.

show abstract

“…We relied on the python scripts and instructions provided by Lau et al (2014) and chose a topic cardinality (i.e., number of top words) of 10 for the calculation. Following explanations by Hao et al (2018), NPMI scores are then first calculated individually for each language and averaged subsequently per model. The higher the NPMI value, the better.…”

Section: Coherence Metricsmentioning

confidence: 99%

Building the Bridge: Topic Modeling for Comparative Research

Lind

Eberl

Eisele

et al. 2021

Communication Methods and Measures

View full text Add to dashboard Cite

In communication research, topic modeling is primarily used for discovering systematic patterns in monolingual text corpora. To advance the usage, we provide an overview of recently presented strategies to extract topics from multilingual text collections for the purpose of comparative research. Moreover, we discuss, demonstrate, and facilitate the usability of the "Polylingual Topic Model" (PLTM) for such analyses. The appeal of this model is that it derives lists of related clustered words in different languages with little reliance on translation or multilingual dictionaries and without the need for manual post-hoc matching of topics. PLTM bridges the gap between languages by making use of document connections in training documents. As these training documents are the crucial resource for the model, we compare model evaluation metrics for different strategies to build training documents. By discussing the advantages and limitations of the different strategies in respect to different scenarios, our study contributes to the methodological discussion on automated content analysis of multilingual text corpora.

show abstract

Lessons from the Bible on Modern Topics: Low-Resource Multilingual Topic Model Evaluation

Cited by 11 publications

References 32 publications

An Empirical Study on Crosslingual Transfer in Probabilistic Topic Models

An Empirical Study on Crosslingual Transfer in Probabilistic Topic Models

Evaluating Topic Quality with Posterior Variability

Building the Bridge: Topic Modeling for Comparative Research

Contact Info

Product

Resources

About