A biterm topic model for short texts

By Xiaohui Yan, Jiafeng Guo, Yanyan Lan, Xueqi Cheng. The Web Conference, 2013

1,119 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/b7da93b0c797cdb2b5b31adc73224d625c0a0759

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

Citations in the corpus

Improving Topic Models with Latent Feature Word Representations (2015, methodology): “After using a topic model to calculate the topic probabilities of a document, we assign every document the topic with the highest probability given the document (Cai et al., 2008; Lu et al., 2011; Xie and Xing, 2013; Yan et al., 2013).”
Mechanism-Aware Neural Machine for Dialogue Response Generation (2017, methodology): “Here, we ﬁrstly train Biterm topic model (Yan et al. 2013) on about 20 million sentences, apply it (with 500 topics) to inferring the topics on the whole corpus, and assign to each response only one topic with the maximal value of topic distribution.”
Topic Modeling for Short Texts with Auxiliary Word Embeddings (2016, methodology): “Although BTM was shown to perform better than DMM in [39], as we discussed earlier, the NB based document representation was used for DMM in [39], which may be the reason for the poorer performance.”
BTM: Topic Modeling over Short Texts (2014, methodology): “Besides, two recently proposed models, i.e., the regularized topic model [20] and the generalized P olya model [21], share the same idea of utilizing word co-occurrence (i.e., biterm) statistics to enhance topic learning.”
Using Topic Modeling Methods for Short-Text Data: A Comparative Analysis (2020, methodology): “Yan et al. (2013) developed a short-text TM method called biterm topic model (BTM) that uses word correlations or embedding to FIGURE 1 | The steps involved in a text mining process (Kaur and Singh, 2019). advance TM.”
Latent Dirichlet allocation (LDA) and topic modeling: models, applications, a survey (2017, background): “The main advantage of BTM prevents the data sparsity issue with learning a global topic distribution [24, 177].”
Approximating Interactive Human Evaluation with Self-Play for Open-Domain Dialog Systems (2019, background): “Given research highlighting the ineffectiveness of LDA for short texts [31], such as those involved in casual conversation, and the unavailability of topic and tone supervision at-scale, approaches overcoming these limitations are preferred.”
Personalized Reason Generation for Explainable Song Recommendation (2019, methodology): “In the second part, we use a biterm language model [60], which counts word pairs separated by zero or more words, to measure skip n-grams in the generated text.”
Semantic Clustering and Convolutional Neural Network for Short Text Categorization (2015, background): “Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Short Papers), pages 352–357, Beijing, China, July 26-31, 2015. c©2015 Association for Computational Linguistics”
Quantifying the Visual Concreteness of Words and Topics in Multimodal Datasets (2018, background): “, 2003) (256 topics, trained with Mallet (McCallum, 2002)) a specialized biterm topic model (BTM) (Yan et al., 2013) for short texts (30 topics), and paragraph vectors (PV) (Le and Mikolov, 2014) (PV-DBOW version, 256 dimensions, trained with Gensim (Řehůřek and Sojka, 2010)).”
An Unsupervised Neural Attention Model for Aspect Extraction (2017, methodology): “(4) BTM (Yan et al., 2013): This is a biterm topic model that is specially designed for short texts such as texts from social media and review sites.”