This chapter covers:
- Building and working with knowledge graphs
- Implementing open information extraction to generate knowledge graphs from text
- Using semantic knowledge graphs for query expansion and rewriting, relationship discovery, query classification, and query-sense disambiguation
- Discovering the nuanced context for interpreting each query
- Interpreting documents with semantic knowledge graphs to power content-based document recommendations
In the last chapter we focused on learning relationships between documents based upon crowdsourced interactions linking those documents. While these interactions were primarily user behavioral signals, we closed out the chapter also discussing links between documents that appear within the documents themselves - for example, leveraging hyperlinks between documents. In chapter 2, we also discussed how the textual content in the documents, rather than being "unstructured data", is really more like a giant graph of "hyperstructured data" containing a rich graph of semantic relationships connecting the many character sequences, terms, and phrases that exist across the fields within our collections of documents.