Device, method and program for natural language processing
Disclosed is a natural language processing technique according to a neural network of high interpretive ability. One embodiment of the present disclosure relates to an apparatus including a trained neural network into which first natural language text is input and that is trained to output second natural language text and alignment information, the second natural language text being in accordance with a predetermined purpose corresponding to the first natural language text, and the alignment information indicating, for each part of the second natural language text, which part of the first natural language text is a basis of information for generation; and an analyzing unit configured to output, upon input text being input into the trained neural network, a predicted result of output text in accordance with a predetermined purpose, and alignment information indicating, for each part of the predicted result of the output text, which part of the input text is a basis of information for generation.
1. An apparatus comprising:
a hardware processor that:
receives from a trained neural network system, a predicted result of output text corresponding to input text and alignment information indicating which part of the input text is a basis of information for generation with respect to the predicted result; and
outputs the predicted result of output text corresponding to the input text and the alignment information,
wherein the trained neural network system, into which first natural language text is input, outputs second natural language text and alignment information, the second natural language text corresponding to the first natural language text, and the alignment information indicating, with respect to the second natural language text, which part of the first natural language text is a basis of information for generation.
2. A non-transitory computer-readable recording medium having a program embodied therein causing a processor to read and execute the trained neural network system of the apparatus according to claim 1 .
3. The apparatus of claim 1 , wherein the alignment information indicates correspondence between substrings in the input text and substrings in the output text.
4. A system comprising:
a hardware processor that, when reading and executing a trained neural network system:
receives from the trained neural network system, a predicted result of output text corresponding to input text and alignment information indicating which part of the input text is a basis of information for generation with respect to the predicted result;
with respect to each of learning data given beforehand consisting of pairs of the input text for learning and correct output text for learning; and
updates each parameter of the trained neural network system, in accordance with a value of a loss function calculated based on the predicted result of the output text for learning and the alignment information,
wherein the trained neural network system, into which first natural language text is input, outputs second natural language text and alignment information, the second natural language text corresponding to the first natural language text, and the alignment information indicating, with respect to the second natural language text, which part of the first natural language text is a basis of information for generation.
5. A non-transitory computer-readable recording medium having a program embodied therein causing a processor to read and execute the trained neural network system of the system of claim 4 .
6. The system of claim 4 , wherein the alignment information indicates correspondence between substrings in the input text and substrings in the output text.
7. A system comprising:
a hardware processor that, when reading and executing a trained neural network system:
receives from the trained neural network system, a predicted result of output text corresponding to input text and alignment information,
indicating which part of the input text is a basis of information for generation with respect to the predicted result,
wherein the trained neural network system, into which first natural language text is input, outputs second natural language text and alignment information, the second natural language text corresponding to the first natural language text, and the alignment information indicating, with respect to the second natural language text, which part of the first natural language text is a basis of information for generation.
8. The system of claim 7 , wherein the alignment information indicates correspondence between substrings in the input text and substrings in the output text.
9. A non-transitory computer-readable recording medium having a program embodied therein causing a processor to read and execute the trained neural network system of the system of claim 7 .