Method for generating question answering robot and computer device

1. A method for generating a question answering robot, comprising:

obtaining field information input by a user, and obtaining a field-specific robot from a robot library based on the field information;

obtaining a template list corresponding to the field-specific robot, and providing the template list to the user, the template list comprising a plurality of templates;

receiving the plurality of templates filled in by the user, the plurality of templates filled in by the user comprising at least one question and an answer corresponding to the at least one question;

expanding the at least one question filled in by the user based on a question semantic database to form a combination of questions corresponding to the answer, the answer and the combination of questions forming a question-answer pair; and

generating a question answering robot based on the question-answer pair;

wherein the field-specific robot is generated by:

obtaining a plurality of existing question answering robots, and obtaining existing question-answer pairs of each existing question answering robot;

generating a field label of each existing question answering robot based on the existing question-answer pairs of the existing question answering robot; and

clustering, based on a plurality of field labels of the plurality of existing question answering robots, question answering robots belonging to a same field to form the field-specific robot of the same field;

wherein the plurality of templates in the template list are generated by:

storing existing question-answer pairs with similar forms of the plurality of existing question-answering robots in a same cluster;

determining an edit distance of any two existing question-answer pairs with different semantic meanings in the same cluster; and

performing alignment processing on any two existing question-answer pairs with different semantic meanings and having a minimum edit distance in the same cluster to obtain the template of the two existing question-answer pairs.

2. The method of claim 1 , wherein clustering, based on the plurality of field labels of the plurality of existing question answering robots, the question answering robots belonging to the same field to form the field-specific robot of the same field comprises:

determining semantic meanings of the existing question-answer pairs of each existing question answering robot;

storing existing question-answer pairs with the same semantic meaning of each existing question answering robot in a same question semantic database; and

merging existing question-answer pairs with the same semantic meaning in different question semantic databases to form the field-specific robot.

3. The method of claim 1 , wherein expanding the at least one question filled in by the user based on the question semantic database comprises:

determining keywords that the user fills in the plurality of templates; and

expanding the at least one question filled in by the user based on the keywords and question-answer pairs stored in the question semantic database.

4. The method of claim 1 , wherein obtaining the field-specific robot from the robot library based on the field information comprises:

determining a field corresponding to the field information based on a question classifier; and

obtaining the field-specific robot from the robot library based on the field corresponding to the field information.

5. A computer device, comprising:

at least one processor; and

a storage device communicatively connected to the at least one processor; wherein,

the storage device stores an instruction executable by the at least one processor, and when the instruction is executed by the at least one processor, the at least one processor may implement the method for generating the question answering robot comprising:

obtaining field information input by a user, and obtaining a field-specific robot from a robot library based on the field information;

obtaining a template list corresponding to the field-specific robot, and providing the template list to the user, the template list comprising a plurality of templates;

receiving the plurality of templates filled in by the user, the plurality of templates filled in by the user comprising at least one question and an answer corresponding to the at least one question;

expanding the at least one question filled in by the user based on a question semantic database to form a combination of questions corresponding to the answer, the answer and the combination of questions forming a question-answer pair; and

generating a question answering robot based on the question-answer pair;

wherein the field-specific robot is generated by:

obtaining a plurality of existing question answering robots, and obtaining existing question-answer pairs of each existing question answering robot;

generating a field label of each existing question answering robot based on the existing question-answer pairs of the existing question answering robot; and

clustering, based on a plurality of field labels of the plurality of existing question answering robots, question answering robots belonging to a same field to form the field-specific robot of the same field;

wherein the plurality of templates in the template list are generated by:

storing existing question-answer pairs with similar forms of the plurality of existing question-answering robots in a same cluster;

determining an edit distance of any two existing question-answer pairs with different semantic meanings in the same cluster; and

performing alignment processing on any two existing question-answer pairs with different semantic meanings and having a minimum edit distance in the same cluster to obtain the template of the two existing question-answer pairs.

6. The device of claim 5 , wherein clustering, based on the plurality of field labels of the plurality of existing question answering robots, the question answering robots belonging to the same field to form the field-specific robot of the same field comprises:

determining semantic meanings of the existing question-answer pairs of each existing question answering robot;

storing existing question-answer pairs with the same semantic meaning of each existing question answering robot in a same question semantic database; and

merging existing question-answer pairs with the same semantic meaning in different question semantic databases to form the field-specific robot.

7. The device of claim 5 , wherein expanding the at least one question filled in by the user based on the question semantic database comprises:

determining keywords that the user fills in the plurality of templates; and

expanding the at least one question filled in by the user based on the keywords and question-answer pairs stored in the question semantic database.

8. The device of claim 5 , wherein obtaining the field-specific robot from the robot library based on the field information comprises:

determining a field corresponding to the field information based on a question classifier; and

obtaining the field-specific robot from the robot library based on the field corresponding to the field information.

9. A non-transitory computer-readable storage medium having a computer instruction stored thereon, wherein the computer instruction is configured to make a computer implement the method for generating the question answering robot comprising:

obtaining field information input by a user, and obtaining a field-specific robot from a robot library based on the field information;

obtaining a template list corresponding to the field-specific robot, and providing the template list to the user, the template list comprising a plurality of templates;

receiving the plurality of templates filled in by the user, the plurality of templates filled in by the user comprising at least one question and an answer corresponding to the at least one question;

expanding the at least one question filled in by the user based on a question semantic database to form a combination of questions corresponding to the answer, the answer and the combination of questions forming a question-answer pair; and

generating a question answering robot based on the question-answer pair;

wherein the field-specific robot is generated by:

obtaining a plurality of existing question answering robots, and obtaining existing question-answer pairs of each existing question answering robot;

generating a field label of each existing question answering robot based on the existing question-answer pairs of the existing question answering robot; and

clustering, based on a plurality of field labels of the plurality of existing question answering robots, question answering robots belonging to a same field to form the field-specific robot of the same field;

wherein the plurality of templates in the template list are generated by:

storing existing question-answer pairs with similar forms of the plurality of existing question-answering robots in a same cluster;

determining an edit distance of any two existing question-answer pairs with different semantic meanings in the same cluster; and

performing alignment processing on any two existing question-answer pairs with different semantic meanings and having a minimum edit distance in the same cluster to obtain the template of the two existing question-answer pairs.

10. The medium of claim 9 , wherein clustering, based on the plurality of field labels of the plurality of existing question answering robots, the question answering robots belonging to the same field to form the field-specific robot of the same field comprises:

determining semantic meanings of the existing question-answer pairs of each existing question answering robot;

storing existing question-answer pairs with the same semantic meaning of each existing question answering robot in a same question semantic database; and

merging existing question-answer pairs with the same semantic meaning in different question semantic databases to form the field-specific robot.

11. The medium of claim 9 , wherein expanding the at least one question filled in by the user based on the question semantic database comprises:

determining keywords that the user fills in the plurality of templates; and

expanding the at least one question filled in by the user based on the keywords and question-answer pairs stored in the question semantic database.

12. The medium of claim 9 , wherein obtaining the field-specific robot from the robot library based on the field information comprises:

determining a field corresponding to the field information based on a question classifier; and

obtaining the field-specific robot from the robot library based on the field corresponding to the field information.