Determination method and determination apparatus
A determination method executed by a computer including a memory and a processor coupled to the memory, includes receiving a plurality of sentences and designation of terms included in the plurality of sentences, generating, for each term for which the designation is received, information indicating a relation between the term and each of other terms included in one of the plurality of sentences containing the term, extracting, for each term for which the designation is received, information indicating a specific relation from the generated information indicating the relation, generating characteristic information that uses the extracted information as a feature, and determining similarity between a plurality of the terms based on the generated characteristic information for each term.
1. A computer-implemented determination method comprising:
receiving a plurality of sentences and designation of terms included in the plurality of sentences;
generating, for each term of a plurality of terms designated by the designation, information indicating relations between each term of plurality of terms and other terms included in each sentence of the plurality of sentences;
extracting, from the generated information, first specific information indicating a first specific relation in the relations of a first term included in a first sentence and second specific information indicating a second specific relation in the relations of a second term included in a second sentence;
generating a first characteristic vector based on the first specific information and a second characteristic vector based on the second specific information;
generating a first specific vector by comparing the first characteristic vector with the second characteristic vector; and
determining similarity between another plurality of the terms on the basis of the first characteristic vector, the second characteristic vector, and the generated first specific vector.
2. The determination method according to claim 1 , wherein the relations is identified by meaning connections between each term of plurality of terms and the other terms.
3. The determination method according to claim 2 , wherein
the first specific relation is extracted by extracting, from the meaning connections of the first term, a meaning connection between the first term and another term which is directly connected to the first term in the meaning connections of the first term.
4. The determination method according to claim 1 , comprising:
generating a second specific vector by synthesizing the first characteristic vector, the second characteristic vector, and the generated first vector,
wherein the determining of the similarity is executed on the basis of the second specific vector.
5. The determination method according to claim 1 , further comprising:
receiving similarity-dissimilarity information indicating whether each term of the plurality of terms is similar to each other; and
performing machine learning by using the first characteristic vector, the second characteristic vector, the generated first specific vector, and the similarity-dissimilarity information,
wherein the determining of the similarity includes determining the similarity between the other plurality of the terms on the basis of a result of a result of the machine learning.
6. The determination method according to claim 4 , wherein
the determining of the similarity includes determining the similarity between the other plurality of the terms on the basis of a result of machine learning executed by using the second specific vector.
7. A determination apparatus comprising:
a memory; and
a processor coupled to the memory and configured to:
receive a plurality of sentences and designation of terms included in the plurality of sentences,
generate, for each term of a plurality of terms designated by the designation, information indicating a relations between each term of plurality of terms and other terms included in each sentence of the plurality of sentences,
extract, from the generated information, first specific information indicating a first specific relation in the relations of a first term included in a first sentence and second specific information indicating a second specific relation in the relations of a second term included in a second sentence,
generate a first characteristic vector based on the first specific information and a second characteristic vector based on the second specific information,
generating a first specific vector by comparing the first characteristic vector with the second characteristic vector, and
determine similarity between another plurality of the terms on the basis of the first characteristic vector, the second characteristic vector, and the generated first specific vector.
8. The determination apparatus according to claim 7 , wherein
the relations is identified by meaning connections between each term of plurality of terms and the other terms.
9. The determination apparatus according to claim 8 , wherein
the first specific relation is extracted by extracting, from the meaning connections of the first term, a meaning connection between the first term and another term which is directly connected to the first term in the meaning connections of the first term.
10. The determination apparatus according to claim 7 , comprising:
generating a second specific vector by synthesizing the first characteristic vector, the second characteristic vector, and the generated first vector,
wherein the determining of the similarity is executed on the basis of the second specific vector.
11. The determination apparatus according to claim 7 , further comprising:
receiving similarity-dissimilarity information indicating whether each term of the plurality of terms is similar to each other; and
performing machine learning by using the first characteristic vector, the second characteristic vector, the generated first specific vector, and the similarity-dissimilarity information,
wherein the determining of the similarity includes determining the similarity between the other plurality of the terms on the basis of a result of a result of the machine learning.
12. The determination apparatus according to claim 10 , wherein
the determining of the similarity includes determining the similarity between the other plurality of the terms on the basis of a result of machine learning executed by using the second specific vector.
13. A non-transitory computer-readable medium storing a determination program executable by one or more computers, the determination program comprising instructions for:
receiving a plurality of sentences and designation of terms included in the plurality of sentences;
generating, for each term of a plurality of terms designated by the designation, information indicating relations between each term of plurality of terms and other terms included in each sentence of the plurality of sentences;
extracting, from the generated information, first specific information indicating a first specific relation in the relations of the first term included in a first sentence and second specific information indicating a second specific relation in the relations of a second term included in a second sentence;
generating a first characteristic vector based on the first specific information and a second characteristic vector based on the second specific information;
generating a first specific vector by comparing the first characteristic vector with the second characteristic vector; and
determining similarity between another plurality of the terms on the basis of the first characteristic vector, the second characteristic vector, and the generated first vector.