Tree index of LlamaIndex

---

Terminology

Node: Documents will be parsed into nodes.

Response Synthesis: This is the process of merging the retrieved nodes into a response output.

Summary Index

In general, we may want to store nodes as a sequential chain, this is called summery index:

During query, if no other query parameters are specified, LlamaIndex will load all nodes into response synthesis module.

From an embedding based query, Summery index can also fetch the top-k neighbors or with the addition of a keyword filter:

Vector Store

This storage method is more efficient when you directly query the embedding top-k.

Tree Index

Keyword Table Index

The keyword table index extracts keywords from each Node and builds a mapping from each keyword to the corresponding Nodes of that keyword.