abstract |
A method, system and computer program product for using a natural language processor are disclosed. Included are importing highlighted and non-highlighted training text, each including training nodes, encoding the training text 1-out-of-n, training a projection model by using the training text, processing the highlighted training text by using the projection model, and training a classifier model by using the highlighted processed training text. Also included are importing new text comprising new nodes, encoding the new text 1-out-of-n, processing the new text using the projection model, and determining whether any of the new nodes have changed using the classifier model is in a searched class. |