The Swedish startup Saplo has released a new text analysis API today that should help developers build tools to tap into the company’s semantic technology platform. This should allow people to build their own personalization or recommendation apps based on contextual meaning rather than solely on tags or keywords.
The API offers access to 5 different services: entity tagging, topic tags, related and similar articles, contextual recognition, and sentiment anaysis.
Sponsor
“For example you can create a resource for a user and, based on what they have liked or shared, recommend other users with the same interest that they should follow. In this way it is possible to find things that usually don’t appear in your social graph but is still highly relevant for you”, says Mattias Tyrberg, Saplo’s CEO.
Saplo’s API is language independent and new languages can be set up automatically. Currently English and Swedish are in production, but Saplo says it has customers who’ve used both Danish and Polish for their recommendations. “All we need is a large set of texts for the language (e.g. Wikipedia articles) to make the initial setup,” says Tyrberg, “and then of course someone who knows the language that can evaluate the results being produced.”
In addition to the new functionalitly in Saplo’s API, there are a number of improvements to the documentation and to the libraries. There are 2000 free calls for developers (for non commercial use).
0 Responses
Stay in touch with the conversation, subscribe to the RSS feed for comments on this post.