Amgad Madkour is a member of the Microsoft Bing knowledge graph team. He has a Ph.D. in computer science from Purdue University where he focused on creating scalable query processing techniques over web-scale graphs. Before joining Microsoft, Amgad was a staff R&D engineer at IBM. At Microsoft, he focuses on developing techniques for constructing, exploring and mining knowledge graphs
2021 Talk: Entity Life Cycle In Search-Centric Knowledge Graphs
Entity-based results are becoming an integral part of the search experience. Search-centric companies highly rely on knowledge graphs in providing the necessary information for building rich search experiences. An entity can originate from a structured, semi-structured, or unstructured data source. An entity passes through a series of stages to be onboarded into a knowledge graph and then served as part of a search result. This talk presents the life cycle of an entity including extraction, schema mapping, ingestion, entity resolution, data quality, and publishing. It covers the challenges and approaches employed at each stage. It also introduces some of the different experiences that can be generated for a given entity. Finally, it covers the multilingual aspect of knowledge graphs and the challenges of maintaining them at scale.