Stanford CS224W: Machine Learning with Graphs | 2021 | Lecture 11.2 – Answering Predictive Queries

For more information about Stanford’s Artificial Intelligence professional and graduate programs, visit: https://stanford.io/3BsESs7

Lecture 11.2 – Answering Predictive Queries on Knowledge Graphs
Jure Leskovec
Computer Science, PhD

We introduce the idea of traversing knowledge graphs in vector space for better reasoning and handling queries. We show how composability relationships can allow us to handle path queries. We then finally talk about how to traverse knowledge graphs to answer conjunctive queries, as well as how to use embeddings to impute missing relationships.

To follow along with the course schedule and syllabus, visit:
http://web.stanford.edu/class/cs224w/