arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2006.12414cs.CL
收藏

用于低资源设置的Dirichlet平滑词嵌入

Dirichlet-Smoothed Word Embeddings for Low-Resource Settings

Center for Information and Language Processing(信息与语言处理中心);Ludwig-Maximilians-Universität München(慕尼黑大学)

Jakob Jungmaier, Nora Kassner, Benjamin Roth

摘要:

如今,使用正逐点互信息(PPMI)加权共生矩阵的经典基于计数的词嵌入已被基于机器学习的方法(如word2vec和GloVe)广泛取代。但是这些方法通常使用非常大量的文本数据来应用。然而,在许多情况下,没有太多的文本数据可用,例如特定领域或低资源语言。本文通过加入Dirichlet平滑来修正其对稀有词的偏见,从而重新审视了PPMI。我们在标准的词相似性数据集上进行评估,并与word2vec和最新的低资源设置的最新技术进行比较:词嵌入的正和未标记(PU)学习。该方法在低资源环境下的性能优于PU学习,并在马耳他语和匈牙利语中获得了有竞争力的结果。

英文摘要:

Nowadays, classical count-based word embeddings using positive pointwise mutual information (PPMI) weighted co-occurrence matrices have been widely superseded by machine-learning-based methods like word2vec and GloVe. But these methods are usually applied using very large amounts of text data. In many cases, however, there is not much text data available, for example for specific domains or low-resource languages. This paper revisits PPMI by adding Dirichlet smoothing to correct its bias towards rare words. We evaluate on standard word similarity data sets and compare to word2vec and the recent state of the art for low-resource settings: Positive and Unlabeled (PU) Learning for word embeddings. The proposed method outperforms PU-Learning for low-resource settings and obtains competitive results for Maltese and Luxembourgish.

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。

↑