Method and apparatus for automatically discovering medical knowledge

1. A computer-implemented method for automatically discovering medical knowledge comprising: obtaining, by at least one processor, one or more linking concepts having a semantic relation with a starting concept from a medical literature library, wherein the starting concept represents a disease; obtaining, by at the least one processor, one or more target concepts having a semantic relation with the one or more linking concepts from the medical literature library; calculating, by at the least one processor, an association degree of each of the one or more target concepts with respect to the starting concept, wherein the association degree indicates a probability that the target concept is able to cope with the starting concept; and sorting, by at the least one processor, the one or more target concepts according to the calculated association degrees; and

displaying the sorting of the target concepts,

wherein calculating the association degree of each of the one or more target concepts with respect to the starting concept comprises: for each of the one or more target concepts,

determining, by the at least one processor, a linking concept related to the target concept in the one or more linking concepts as a related linking concept and

calculating, by the at least one processor, the association degree of the target concept with respect to the starting concept based on a first semantic relation between the related linking concept and the starting concept and a second semantic relation between the related linking concept and the target concept by using a pretrained Markov logic network, wherein the Markov logic network is composed of predefined predicates and logic formulas describing a logic relation among the predicates.

2. The method according to claim 1 , wherein the association degree of the target concept with respect to the starting concept is calculated as below:

P

=

1

Z

⁢

exp

⁡

(

Σ

i

⁢

⁢

w

i

⁢

f

i

)

wherein P represents the association degree, Z represents a normalization factor, f i represents the i th logic formula, and w i represents a weight for the i th logic formula.

3. The method according to claim 1 , further comprising: providing a logic relation between each of the one or more target concepts and the starting concept.

4. The method according to claim 3 , wherein providing a logic relation between each of the one or more target concepts and the starting concept comprises: for each of the one or more target concepts, determining a logic formula including the first semantic relation and the second semantic relation as the logic relation; and recording the logic relation in association with the target concept.

5. The method according to claim 1 , wherein obtaining one or more linking concepts having the semantic relation with the starting concept from the medical literature library comprises: retrieving a sentence containing the starting concept from the medical literature library;

extracting the semantic relation contained in the sentence; and determining the one or more linking concepts based on the semantic relation.

6. The method according to claim 5 , wherein obtaining one or more linking concepts having the semantic relation with the starting concept from the medical literature library further comprises: filtering the obtained linking concepts to obtain the linking concepts having a predetermined semantic relation.

7. The method according to claim 1 , wherein obtaining one or more target concepts having the semantic relation with the one or more linking concepts from the medical literature library comprises: for each of the one or more linking concepts, retrieving a sentence containing the linking concept from the medical literature library; extracting the semantic relation contained in the sentence; and determining the target concept based on the semantic relation.

8. The method according to claim 1 , wherein the one or more target concepts are sorted according to a descending order of the corresponding association degrees.

9. An apparatus for automatically discovering medical knowledge comprising: at least one processor; and at least one memory storing a computer program; wherein when the computer program is executed by the at least one processor, the apparatus is caused to:

obtain one or more linking concepts having a semantic relation with a starting concept from a medical literature library, wherein the starting concept represents a disease; obtain one or more target concepts having a semantic relation with the one or more linking concepts from the medical literature library; calculate an association degree of each of the one or more target concepts with respect to the starting concept, wherein the association degree indicates a probability that the target concept is able to cope with the starting concept; and sort the one or more target concepts according to the calculated association degrees,

wherein the apparatus further comprise a display device configured to display the sorting of the target concepts, and

wherein when the computer program is executed by the at least one processor, the apparatus is caused to calculate the association degree of each of the one or more target concepts with respect to the starting concept by the following operations: for each of the one or more target concepts,

determining a linking concept related to the target concept in the one or more linking concepts as a related linking concept; and

calculating the association degree of the target concept with respect to the starting concept based on a first semantic relation between the related linking concept and the starting concept and a second semantic relation between the related linking concept and the target concept by using a pretrained Markov logic network, wherein the Markov logic network is composed of predefined predicates and logic formulas describing a logic relation among the predicates.

10. The apparatus according to claim 9 , wherein the association degree of the target concept with respect to the starting concept is calculated as below:

P

=

1

Z

⁢

exp

⁡

(

Σ

i

⁢

⁢

w

i

⁢

f

i

)

wherein P represents the association degree, Z represents a normalization factor, f i represents the i th logic formula, and w i represents a weight for the i th logic formula.

11. The apparatus according to claim 9 , wherein when the computer program is executed by the at least one processor, the apparatus is further caused to provide a logic relation between each of the one or more target concepts and the starting concept.

12. The apparatus according to claim 11 , wherein when the computer program is executed by the at least one processor, the apparatus is caused to provide a logic relation between each of the one or more target concepts and the starting concept by the following operations: for each of the one or more target concepts, determining a logic formula including the first semantic relation and the second semantic relation as the logic relation; and recording the logic relation in association with the target concept.

13. The apparatus according to claim 9 , wherein when the computer program is executed by the at least one processor, the apparatus is caused to obtain one or more linking concepts having the semantic relation with the starting concept from the medical literature library by the following operations: retrieving a sentence containing the starting concept from the medical literature library; extracting the semantic relation contained in the sentence; and determining the one or more linking concepts based on the semantic relation.

14. The apparatus according to claim 13 , wherein when the computer program is executed by the at least one processor, the apparatus is further caused to obtain one or more linking concepts having the semantic relation with the starting concept from the medical literature library by the following operation: filtering the obtained linking concepts to obtain the linking concepts having a predetermined semantic relation.

15. The apparatus according to claim 9 , wherein when the computer program is executed by the at least one processor, the apparatus is caused to obtain one or more target concepts having the semantic relation with the one or more linking concepts from the medical literature library by the following operations: for each of the one or more linking concepts, retrieving a sentence containing the linking concept from the medical literature library; extracting the semantic relation contained in the sentence; and determining the target concept based on the semantic relation.

16. The apparatus according to claim 9 , wherein when the computer program is executed by the at least one processor, the apparatus is caused to sort the one or more target concepts according to a descending order of the corresponding association degrees.

17. A computer readable non-transitory storage medium storing a computer program, wherein when the computer program is executed by a processor, the steps of the method for automatically discovering medical knowledge according to claim 1 are carried out.

18. The method according to claim 2 , further comprising: providing a logic relation between each of the one or more target concepts and the starting concept.