Asset Management Using Knowledge Graphs
Originally Aired - Monday, April 17 | 3:40 PM - 4:00 PM PT
This Session Has Not Started Yet
Be sure to come back after the session starts to have access to session resources.
Broadcast studios need to manage thousands of assets from hundreds of different manufacturers. Knowing which firmware update is applicable to which asset is becoming increasingly challenging and can cause expensive asset downtime if not done properly. Many studios have incomplete, or inaccurate asset descriptions, typically described in databases or spreadsheets, which complicate asset identification. Luckily, asset descriptions can be found in many other openly accessible data sources including OEM Web pages, and online retail Websites. These additional data sources can be used to uniquely identify each asset, as well as to create a more complete asset description, including information about updates and warranty. Knowledge graphs are useful as a universal data format for asset descriptions such that asset descriptions of two different data sources can be easily compared. Knowledge graphs also provide powerful querying capabilities, and can form the basis for advanced graph-based machine learning to detect complex patterns related to asset compatibility, or asset malfunction.
In this talk, we want to show how a knowledge graph can be used to improve asset management.