SpringSense Meaning Recognition API v2.0 Retakes World’s Best Accuracy in Industry Benchmark
MELBOURNE, Australia, May 8, 2013 /PRNewswire/ — SpringSense is once again the world’s most accurate noun-sense disambiguator, surpassing the latest in academic research to regain the top position. SpringSense remains the only Natural Language Processing (NLP) solution able to process English text in real time with near human accuracy, making it perfectly suited to Big Data and other high volume applications.
The latest version of the SpringSense Meaning Recognition API is now even more accurate with a score of 83.4% in the industry benchmark SemEval 4/7.
The patent-pending v2.0 API is available now for commercial use, with a free trial, via the Mashape API Hub.
Requiring no external interaction and able to process large amounts of text in real-time, SpringSense is ideal for high-volume transactional applications such as Enterprise Search and Big Data, in contrast to the previous accuracy leader, an academic solution which is not able to perform recognition in real-time. SpringSense owes its speed to an innovative approach to NLP which uses a patent-pending data-mining algorithm.
The benchmark used, SemEval, is an ongoing series of evaluations of computational semantic analysis systems, used as the standard method of quantifying the accuracy of Natural Language Processing algorithms.
The full story of how SpringSense regained its position as the world’s most accurate noun-sense disambiguator can be found at http://springsense.com/news/
SpringSense is a young Australian startup dedicated to delivering our patent-pending natural-language processing algorithm to a global audience. SpringSense is a subsidiary of DiUS, a leading Australian software technology services company with offices nationally. www.springsense.com
Mashape is the Cloud API Hub which provide a world-class marketplace to manage, distribute and consume any kind of API in the world, both cloud and internal, both existing or just born, targeting every developer and organization committed into the internet.
The API is available at: