Scientific Formula Retrieval via Tree Embeddings
Scientific Formula Retrieval via Tree Embeddings
复制标题
DOI：
10.1109/bigdata52589.2021.9671942
复制
发表时间：
2021-12
期刊：
影响因子：
--
通讯作者：
Zichao Wang;Mengxue Zhang;Richard Baraniuk;Andrew S. Lan
中科院分区：
文献类型：
--
作者：
Zichao Wang;Mengxue Zhang;Richard Baraniuk;Andrew S. Lan
利用不断增长的科学内容语料库需要新的方法和方法来有效地组织、搜索和检索科学公式。我们提出了一种新的数据驱动框架，用于通过基于树嵌入的学习公式表示来检索相似的科学公式。Forte(用于通过树嵌入的公式表示学习)利用符号科学公式(例如数学公式)的运算符树表示来显式地捕获其固有的结构和语义属性。Forte使用i)树编码器，其将公式的运算符树编码为嵌入向量，以及ii)树解码器，其直接从嵌入向量生成公式的运算符树。我们还提出了一种新的树束搜索算法，提高了解码后的算子树的质量。我们使用一个包含在线收集的770k个科学公式的真实数据集，证明了Forte(有时非常显著)在公式重建和检索方面的性能优于各种基线方法。
Exploiting the ever-growing corpus of scientific content calls for new ways and means to effectively organize, search, and retrieve scientific formulae. We propose a new data-driven framework for retrieving similar scientific formulae via learned formula representations based on tree embeddings. FORTE (for FOrmula Representation learning via Tree Embeddings) leverages operator tree representations of symbolic scientific formulae (such as math equations) to explicitly capture their inherent structural and semantic properties. FORTE employs i) a tree encoder that encodes the formula’s operator tree into an embedding vector and ii) a tree decoder that directly generates a formula’s operator tree from the embedding vector. We also develop a novel tree beam search algorithm that improves the quality of the decoded operator trees. We demonstrate that FORTE (sometimes significantly) outperforms various baseline methods on formula reconstruction and retrieval using a real-world dataset comprising 770k scientific formulae collected on-line.