Metabolomics Publications in PubMed: Embeddings and Associated Datasets

File 1: pubmed_metabolomics_clean_01MAR2024.xml

This is a cleaned-up XML file (entries with errors removed) for PubMed metabolomics-related publications covering the period from 1998 to early 2024.

File 2: metabolomics_embeddings_full_tSNE_uMAP_01MAR2024.h5

This file contains the complete embeddings for metabolomics-related articles from PubMed and associated datasets, spanning from 1998 to early 2024. It includes: 1) PubMed data such as PMID, title, abstract, language, journal title, publication year, and authors. 2) The full 768-dimensional embeddings generated using PubMedBERT. 3) uMAP and t-SNE embeddings.

文件1：pubmed_metabolomics_clean_01MAR2024.xml
本文件为已剔除错误条目的规范化XML格式文件，涵盖1998年至2024年初期间PubMed代谢组学相关学术出版物。

文件2：metabolomics_embeddings_full_tSNE_uMAP_01MAR2024.h5
本文件包含PubMed平台代谢组学相关文章及其关联数据集的完整嵌入向量，时间跨度为1998年至2024年初。其涵盖如下内容：1）PubMed元数据，包括PMID、标题、摘要、语言、期刊名称、发表年份及作者信息；2）采用PubMedBERT生成的768维全量嵌入向量；3）uMAP与t-SNE嵌入向量。