Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[node embedding] node_embedding_algorithm 的疑问 #283

Open
SpikeYangRc opened this issue Nov 15, 2024 · 2 comments
Open

[node embedding] node_embedding_algorithm 的疑问 #283

SpikeYangRc opened this issue Nov 15, 2024 · 2 comments

Comments

@SpikeYangRc
Copy link

SpikeYangRc commented Nov 15, 2024

经过代码分析,我发现虽然系统中声明了 node_embedding_algorithm=node2vec 作为节点嵌入算法,但实际运行中并未调用 _node2vec_embed 进行嵌入计算。在进行 entity、relationship 和 text units 的召回时,系统仅仅使用了图数据库最基础的 get_node、get_edge 等 graph 操作来查询直接关系,这种使用方式完全可以用普通的关系型数据库来实现。那么,使用 graph 的主要目的是否仅仅局限于数据存储和计算 degree 用于排序?值得注意的是,目前的实现并未充分发挥 graph 在处理多级间接关系方面的优势,比如无法实现从 A 经过 B、C 最终到达 D 这样的间接关系路径发现。

@LarFii
Copy link
Collaborator

LarFii commented Nov 19, 2024

非常感谢您的关注,实际上在做检索的过程中我们是利用向量进行的匹配,同时在local信息中也会包含节点的一阶邻居。当然,我们目前确实没有利用graph处理多级间接关系,这也是我们未来探索的方向之一。后续我们也会添加更多的图相关功能,包括自定义KG等等。

@imasoul2
Copy link

Hey @LarFii , would you have a roadmap of when node2vec will be utilised in LightRAG, also additionally, would this have huge improvement over the querying stage of RAG?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants