首页|Patent Issued for Annotating and modeling natural language semantics through ann otation conversion (USPTO 12154552)
Patent Issued for Annotating and modeling natural language semantics through ann otation conversion (USPTO 12154552)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
From the background information supplied by the inventors, news correspondents o btained the followingquote: “The task of natural language understanding (NLU) i nvolves the assignment of semanticallyinterpretable labels to a user’s utteranc e (i.e., textual or spoken expression). These labels can take theform of intent labels, entity labels, or speech/dialog acts, among others. Typically, a corpus of utterancesis annotated with these labels and machine learning models (such as Deep Neural Networks (DNNs)) aretrained on such corpora to predict the label s based on the features extracted from the input sentence.
NETL
NSTL
