[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[EP-tech] Azure AI and EPrints
Just wondering if anyone is experimenting with using the AI services in Microsoft Azure (or any other provider) to generate extra metadata?
I've just spent an experimental half hour or so pasting abstracts from our KAR system here into the online demo for their Text Analytics service. This returns some pretty good looking results. You get back the language it thinks the resource is in, key phrases (which look great for searches), sentiment (probably not that useful in our context), named entities (which look full of possibilities, e.g. you could run it over your repository to generate a map of where your content relates to) and linked entities.
The online demo is here: https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fazure.microsoft.com%2Fen-us%2Fservices%2Fcognitive-services%2Ftext-analytics%2F&data=01%7C01%7Ceprints-tech%40ecs.soton.ac.uk%7C92b2ec9e38cd49b986d408d7c67db5f3%7C4a5378f929f44d3ebe89669d03ada9d8%7C0&sdata=FuQd1vA9%2F9uBZxAv%2BKOxHiCJR%2BCoFc%2FKtDfT1PxZlrE%3D&reserved=0
There are a lot of other services under the same umbrella which also look useful.
Would be very interested to hear if anyone else is experimenting in a similar way and what sort of value you think these services could add.
Liam Green-Hughes | Library Systems Developer
Learning and Research Development, Information Services, University of Kent
-------------- next part --------------
An HTML attachment was scrubbed...