Product Space Networks
收藏DataCite Commons2025-03-31 更新2025-04-15 收录
下载链接：
https://dataverse.harvard.edu/citation?persistentId=doi:10.7910/DVN/FCDZBN
下载链接
链接失效反馈官方服务：
资源简介：
<p>
The Product Space layout is provided in CSV format. The node data (umap_layout_hs92.csv) contains product codes, (X, Y) coordinates, and cluster names. The coordinates represent UMAP embeddings of a proximity matrix, where distances between products are meaningful and represent their relatedness based on whether products are co-exported by countries. Products have also been clustered into knowledge clusters based on co-export patterns.
</p>
<p>
The edge data (top_edges_hs92.csv) is an adjacency list with source and target fields matching the product codes, used to show top relatedness links between products for visualization purposes. Note that the node positions in the node data consider the entire proximity matrix, while the edge data only shows the top relatedness links for visualization purposes. The layout has been optimized for visualization, with uninformative links removed.
</p>
<p>
A data dictionary for this repository describing columns within these files can be found under "product_space_data_dictionary.csv".
</p>
提供机构：
Harvard Dataverse
创建时间：
2019-06-03
