Custom Analyzers, Search Analytics, & Portal Querying in Azure Search
Video | English | Claim
Back in the studio is a man who I think lives at Channel 9, which is awesome because we absolutely love the cool stuff he talks about! ;) Liam Cavanagh is back to talk about Azure Search and the new enhancements to the Azure portal, new search traffic analytics capabilities and customizable analyzers. Liam is awesome, as usual, dropping right in to demo mode, showing off the new . Building off of the Stack Exchange data and information from the last Data Exposed Azure Search video, Liam shows us the ability to do queries right inside the portal in Search Explorer. AWESOME! Liam shows the full capabilities of searching in the Azure portal by executing Lucene queries and other features from the search capabilities.
At the 5:35 mark, we move to a great new feature for doing search analytics by being able to log all the search traffic on your search service. Liam demos how easy it is to enable and configure the Search traffic analytics and the types of information gathered to provide the telemetry data that you can then do analytics on, such as average query latency, number of queries, and more search activity. Liam shows where and how this telemetry data is stored, then at the 9:35 mark Liam shows how to connect to the data via Power BI to get insights into the telemetry search data that was collected to visually look at and get insights into the data. Learn more about this new PowerBI integration...
Finally, at the 11:15 mark Liam goes back to his Postman tool to demonstrate a cool preview feature called Custom Analyzers. Liam first explains how a custom analyzer provides the ability to take full control over the process of breaking your documents into searchable terms to support things such as phonetic search, fast prefix/suffix queries, and much more. This is a very cool demo illustrating how administrators can have a lot more control over how data is analyzed and interpreted. Very cool demo.
© Microsoft Channel 9. Lizenzierung durch die Creative Commons Attribution-Noncommercial-No Derivative Works 3.0-Lizenz.
