arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.06563cs.CLcs.IRcs.LG
收藏

基于预训练主题和词嵌入的神经主题建模中的多视角多源迁移

Multi-view and Multi-source Transfers in Neural Topic Modeling with Pretrained Topic and Word Embeddings

Siemens AG(西门子股份公司);University of Munich (LMU)(慕尼黑大学(LMU))

Pankaj Gupta, Yatin Chaudhary, Hinrich Schütze

摘要:

虽然词嵌入和主题是互补的表示,但过去的一些工作仅在(神经)主题建模中使用预训练的词嵌入来解决短文本或小文档集合中的数据稀疏问题。然而,没有先前的工作采用(预先训练的潜在)的主题在迁移学习范例。在本文中,我们提出了一种方法(1)使用从大型源语料库中获得的潜在主题进行知识转移,以及(2)通过神经主题建模中的两种表示(或视图)联合转移知识,以提高主题质量,更好地处理目标语料库中的多义和数据稀疏问题。在此过程中,我们首先从一个或多个源语料库中积累主题和单词表示,以构建主题和单词向量池。然后,我们识别一个或多个相关的源域,并通过相应的池利用相应的主题和单词特征来指导稀疏目标域中的有意义的学习。我们量化的主题和文件表示的质量,通过泛化(困惑),可解释性(主题连贯性)和信息检索(IR)使用短文本,长文本,小型和大型的文件集,从新闻和医疗领域。我们已经证明了国家的最先进的结果主题建模与建议的框架。

英文摘要:

Though word embeddings and topics are complementary representations, several past works have only used pre-trained word embeddings in (neural) topic modeling to address data sparsity problem in short text or small collection of documents. However, no prior work has employed (pre-trained latent) topics in transfer learning paradigm. In this paper, we propose an approach to (1) perform knowledge transfer using latent topics obtained from a large source corpus, and (2) jointly transfer knowledge via the two representations (or views) in neural topic modeling to improve topic quality, better deal with polysemy and data sparsity issues in a target corpus. In doing so, we first accumulate topics and word representations from one or many source corpora to build a pool of topics and word vectors. Then, we identify one or multiple relevant source domain(s) and take advantage of corresponding topics and word features via the respective pools to guide meaningful learning in the sparse target domain. We quantify the quality of topic and document representations via generalization (perplexity), interpretability (topic coherence) and information retrieval (IR) using short-text, long-text, small and large document collections from news and medical domains. We have demonstrated the state-of-the-art results on topic modeling with the proposed framework.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑