Apparatus and method for verifying sentence
Disclosed are an apparatus and method for verifying a sentence. According to one embodiment of the present disclosure, the method for verifying a sentence includes receiving a sentence automatically generated from a plurality of pieces of data, and the plurality of pieces of data; generating a first embedding vector for each word included in the sentence; generating a second embedding vector for each of the plurality of pieces of data; and generating a verification result value regarding the sentence by using a determination model including a first convolutional neural network for generating a feature vector from the first embedding vector and a second convolutional neural network for generating a feature vector from the second embedding vector.
1. A method for verifying a sentence performed in a computing device comprising one or more processors and a memory configured to store one or more programs to be executed by the one or more processors, the method comprising:
receiving a sentence automatically generated from a plurality of pieces of data, and the plurality of pieces of data;
generating a first embedding vector for each word included in the sentence;
generating a second embedding vector for each of the plurality of pieces of data; and
generating a verification result value regarding the sentence by using a determination model comprising a first convolutional neural network for generating a feature vector from the first embedding vector and a second convolutional neural network for generating a feature vector from the second embedding vector,
wherein the plurality of pieces of data include structured data, and
wherein the verification result value indicates whether or not the sentence is correctly generated from the plurality of pieces of data.
2. The method of claim 1 , wherein the first convolutional neural network comprises a first convolution layer for outputting a first feature vector by performing a convolution operation to the first embedding vector, and a first pooling layer for outputting a second feature vector by performing sub-sampling to the first feature vector, and
the second convolutional neural network comprises a second convolution layer for outputting a third feature vector by performing a convolution operation to the second embedding vector, and a second pooling layer for outputting a fourth feature vector by performing sub-sampling to the third feature vector.
3. The method of claim 2 , wherein the first convolution layer and the second convolution layer perform the convolution operation using a hyperbolic tangent function or a rectified linear unit (ReLU) function as an activation function.
4. The method of claim 2 , wherein the first pooling layer and the second pooling layer perform the sub-sampling using a max pooling function.
5. The method of claim 2 , wherein the determination model further comprises
a fully-connected layer for receiving a fifth feature vector obtained by concatenating the second feature vector and the fourth feature vector; and
an output layer for outputting the verification result value regarding the sentence from output of the fully-connected layer.
6. The method of claim 5 , wherein the output layer outputs the verification result value using a softmax function as an activation function.
7. An apparatus for verifying a sentence, the apparatus comprising:
one or more hardware processors and one or more computer readable media storing instructions that, when executed by the one or more hardware processors, cause the apparatus to:
receive a sentence automatically generated from a plurality of pieces of data, and the plurality of pieces of data;
generate a first embedding vector for each word included in the sentence;
generate a second embedding vector for each of the plurality of pieces of data; and
generate a verification result value regarding the sentence by using a determination model comprising a first convolutional neural network for generating a feature vector from the first embedding vector and a second convolutional neural network for generating a feature vector from the second embedding vector,
wherein the plurality of pieces of data include structured data, and
wherein the verification result value indicates whether or not the sentence is correctly generated from the plurality of pieces of data.
8. The apparatus of claim 7 , wherein the first convolutional neural network comprises a first convolution layer for outputting a first feature vector by performing a convolution operation to the first embedding vector, and a first pooling layer for outputting a second feature vector by performing sub-sampling to the first feature vector, and
the second convolutional neural network comprises a second convolution layer for outputting a third feature vector by performing a convolution operation to the second embedding vector, and a second pooling layer for outputting a fourth feature vector by performing sub-sampling to the third feature vector.
9. The apparatus of claim 8 , wherein the first convolution layer and the second convolution layer perform the convolution operation using a hyperbolic tangent function or a rectified linear unit (ReLU) function as an activation function.
10. The apparatus of claim 8 , wherein the first pooling layer and the second pooling layer perform the sub-sampling using a max pooling function to perform the sub-sampling.
11. The apparatus of claim 8 , wherein the determination model further comprises
a fully-connected layer for receiving a fifth feature vector obtained by concatenating the second feature vector and the fourth feature vector; and
an output layer for outputting the verification result value regarding the sentence from output of the fully-connected layer.
12. The apparatus of claim 11 , wherein the output layer outputs the verification result value using a softmax function as an activation function.