Natural language based service selection system and method, service query system and method

• US 8,117,178 B2
• Filed: 09/26/2008
• Issued: 02/14/2012
• Est. Priority Date: 09/30/2007
• Status: Expired due to Fees

• Current Assignee: NEC Company Limited
• Original Assignee: NEC Company Limited
• Inventors: Qi, Hongwei, Hu, Weisong
• Primary Examiner(s): DANG, THANH HA T
• Application Number:
  US12/239,422
• Publication Number:
  US 20090089282A1
• Time in Patent Office: 1,236 Days
• Field of Search: 707/706, 707/707, 707/708, 707/771
• US Class Current: 707/706
• CPC Class Codes: G06F 16/3344 using natural language analysis

Abstract

The present invention relates to a natural language based service selection system for complementing incomplete queries, which comprises a semantic analyzing device which analyzes an incomplete query from a user semantically, a service selecting device which complements the incomplete query based on the semantic-analyzed query so as to acquire the corresponding selected service, and a retrieving device which retrieves an answer according to the selected service. The present invention also relates to a natural language based service selection to method as well as a service query system and method thereof, and thus can process an incomplete query from a user and provide a selected service.

Claims

1. A natural language based service selection system for complementing incomplete queries, comprising:

a semantic analyzing device which analyzes an incomplete query from a user semantically;

a service selecting device which complements the incomplete query based on a semantically-analyzed query to complete the incomplete query, and selects a corresponding selected service from among a plurality of services based on a complete query;

a retrieving device which retrieves an answer according to the corresponding selected service; and

at least one processor coupled to the semantic analyzing device, the service selecting device and the retrieving device,

wherein the service selecting device comprises a semi-automatic service selecting section which searches lost content in the incomplete query by using a service mapping rule base and complements the lost content through an interacting with the user,

wherein the semi-automatic service selecting section comprises;

a lost content searching unit which matches the semantically-analyzed query with a service mapping rule in the service mapping rule base, and extracts a service type to which the incomplete query belongs and a lost parameter in the incomplete query;

a user interacting unit which prompts the user to input prompt information of a parameter value corresponding to the lost parameter and receives feedback information including the parameter value from the user;

a parameter value extracting unit which extracts the parameter value from the feedback information of the user; and

a query complementing unit which adds the service type, the lost parameter and the parameter value into the semantically-analyzed query to form the complete query.

2. The natural language based service selection system according to claim 1, wherein the lost content searching unit retrieves the service mapping rule meeting the following conditions as matched service mapping rule from the service mapping rule base:

a requirement of the service mapping rule is identical to a requirement of the semantically-analyzed query; and

a service parameter of the service mapping rule contains a parameter of the semantically-analyzed query.

3. The natural language based service selection system according to claim 1, wherein the parameter value extracting unit retrieves the parameter value corresponding to the lost parameter by semantically marking the feedback information of the user.

4. The natural language based service selection system according to claim 1, wherein the service selecting device comprises:

an automatic service selecting section which searches lost content in a current query by using the service mapping rule base and complements the lost content through searching a current user query history base or an other user query history base or a fact base to form the complete query; and

the semi-automatic service selecting section which searches the lost content in the current query by using the service mapping rule base and complements the lost content through interacting with the user to form the complete query, when the corresponding selected service generated by the automatic service selecting section is not accurate.

5. A natural language based service selection system for complementing incomplete queries, comprising:

a semantic analyzing device which analyzes an incomplete query from a user semantically;

a service selecting device which complements the incomplete query based on a semantically-analyzed query to complete the incomplete query, and selects a corresponding selected service from among a plurality of services based on a complete query;

a retrieving device which retrieves an answer according to the corresponding selected service; and

at least one processor coupled to the semantic analyzing device, the service selecting device and the retrieving device,

wherein the service selecting device comprises an automatic service selecting section which searches lost content in a current query by using a service mapping rule base and complements the lost content through searching a current user query history base,

wherein the automatic service selecting section comprises;

a lost content searching unit which matches the semantically-analyzed query with a service mapping rule in the service mapping rule base, and extracts a service type to which the current query belongs and a lost parameter in the current query;

a latest query detecting unit which searches a latest query contained the lost parameter in the current query from the current user query history base, and extracts the parameter value corresponding to the lost parameter in the current query; and

a query complementing unit which adds the service type, the lost parameter and the parameter value into the semantically-analyzed query to form the complete query.

6. The natural language based service selection system according to claim 5, wherein the automatic service selecting section further comprises a similar query detecting unit which searches a history query that contains the lost parameter in the current query as a similar query from the current user query history base when there is no parameter value extracted by the latest query detecting unit, and extracts the parameter value corresponding to the lost parameter in the current query from the similar query.

7. The natural language based service selection system according to claim 6, wherein the similar query detecting unit retrieves the history query meeting the following conditions as a similar query from the current user query history base:

a service type of the history query is identical to a service type to which the current query belongs; and

a query parameter of the history query contains a query parameter of the semantically-analyzed query.

8. The natural language based service selection system according to claim 5, wherein the lost content searching unit retrieves the service mapping rule meeting the following conditions as a matched service mapping rule from the service mapping rule base:

a requirement in the service mapping rule is identical to a requirement of the semantically-analyzed query; and

a service parameter in the service mapping rule contains a parameter of the semantically-analyzed query.

9. A natural language based service selection system for complementing incomplete queries, comprising:

a semantic analyzing device which analyzes an incomplete query from a user semantically;

a service selecting device which complements the incomplete query based on a semantically-analyzed query to complete the incomplete query, and selects a corresponding selected service from among a plurality of services based on a complete query;

a retrieving device which retrieves an answer according to the corresponding selected service; and

at least one processor coupled to the semantic analyzing device, the service selecting device and the retrieving device,

wherein the service selecting device comprises an automatic service selecting section which searches lost content in a current query by using a service mapping rule base and complements the lost content through searching an other user query history base,

wherein the automatic service selecting section comprises;

a lost content searching unit which matches the semantically-analyzed query with a service mapping rule in the service mapping rule base, and extracts a service type to which the current query belongs and a lost parameter in the current query;

a similar query detecting unit which searches a history query that contains the lost parameter in the current query as a similar query from the other user query history base, and extracts a parameter value corresponding to the lost parameter in the current query from the similar query; and

a query complementing unit which adds the service type, the lost parameter and the parameter value into the semantically-analyzed query to form the complete query.

10. The natural language based service selection system according to claim 9, wherein the lost content searching unit retrieves the service mapping rule meeting the following conditions as matched service mapping rule from the service mapping rule base:

a requirement in the service mapping rule is identical to a requirement of the semantically-analyzed query; and

a service parameter in the service mapping rule contains a parameter of the semantically-analyzed query.

11. The natural language based service selection system according to claim 9, wherein the similar query detecting unit retrieves the history query meeting the following conditions as a similar query from the other user query history base:

a service type of the history query is identical to the service type to which the current query belongs; and

a query parameter of the history query contains a query parameter of the semantically-analyzed query.

12. A natural language based service selection system for complementing incomplete queries, comprising:

a semantic analyzing device which analyzes an incomplete query from a user semantically;

a service selecting device which complements the incomplete query based on a semantically-analyzed query to complete the incomplete query, and selects a corresponding selected service from among a plurality of services based on a complete query;

a retrieving device which retrieves an answer according to the corresponding selected service; and

at least one processor coupled to the semantic analyzing device, the service selecting device and the retrieving device,

wherein the service selecting device comprises a third an automatic service selecting section which searches lost content in a current query by using a service mapping rule base and complements the lost content through searching a fact base,

wherein the automatic service selecting section comprises;

a lost content searching unit which matches the semantically-analyzed query with a service mapping rule in the service mapping rule base, and extracts a service type to which the incomplete query belongs and a lost parameter in the incomplete query;

a fact matching unit which matches the semantically-analyzed query with a fact in a fact base to find a matched fact, and extracts a fault value from the matched fact as a lost parameter value; and

a query complementing unit which adds the service type, the lost parameter and the lost parameter value into the semantically-analyzed query to form the complete query.

13. The natural language based service selection system according to claim 12, wherein the lost content searching unit retrieves the service mapping rule meeting the following conditions as a matched service mapping rule from the service mapping rule base:

a requirement in the service mapping rule is identical to a requirement of the semantically-analyzed query; and

a service parameter in the service mapping rule contains a parameter of the semantically-analyzed query.

14. The natural language based service selection system according to claim 12, wherein the fact matching unit retrieves the fact meeting the following conditions as matched fact from the fact base:

a service type of the fact is identical to the service type to which the query belongs; and

a lost parameter in the fact is identical to the lost parameter in the query.

15. A method of selecting service based on an incomplete query, comprising:

a semantic analyzing step, performed by at least one processor, of analyzing an incomplete query from a user semantically;

a service selecting step, performed by the at least one processor, of complementing the incomplete query based on a semantically-analyzed query to complete the incomplete query, and selecting a corresponding selected service from among a plurality of services based on a complete query; and

a retrieving step, performed by the at least one processor, of retrieving an answer according to the corresponding selected service,

wherein the service selecting step comprises a semi-automatic service selecting step of searching lost content in the incomplete query by using a service mapping rule base and complementing the lost content through an interacting with the user,

wherein the semi-automatic service selecting step comprises;

a lost content searching step of matching the semantically-analyzed query with a service mapping rule in the service mapping rule base, and extracting a service type to which the incomplete query belongs and a lost parameter in the incomplete query;

a user interacting step of prompting the user to input prompt information of a parameter value corresponding to the lost parameter and receiving feedback information including the parameter value from the user;

a parameter value extracting step of extracting the parameter value from the feedback information of the user; and

a query complementing step of adding the service type, the lost parameter and the parameter value into the semantically-analyzed query to form the complete query.

16. The method according to claim 15, wherein the lost content searching step comprises a step of retrieving the service mapping rule meeting the following conditions as matched service mapping rule from the service mapping rule base:

a requirement of the service mapping rule is identical to a requirement of the semantically-analyzed query; and

a service parameter of the service mapping rule contains a parameter of the semantically-analyzed query.

17. The method according to claim 15, wherein the parameter value extracting step retrieves the parameter value corresponding to the lost parameter by semantically marking the feedback information of the user.

18. The method according to claim 15, wherein the service selecting step comprises:

an automatic service selecting step of searching lost content in a current query by using the service mapping rule base and complements the lost content through searching a current user query history base or an other user query history base or a fact base to form the complete query; and

the semi-automatic service selecting step of searching the lost content in the current query by using the service mapping rule base and complementing the lost content through interacting with the user to form the complete query, when the corresponding selected service generated by the automatic service selecting step is not accurate.

19. A method of selecting service based on an incomplete query, comprising:

a semantic analyzing step, performed by at least one processor, of analyzing an incomplete query from a user semantically;

a service selecting step, performed by the at least one processor, of complementing the incomplete query based on a semantically-analyzed query to complete the incomplete query, and selecting a corresponding selected service from among a plurality of services based on a complete query; and

a retrieving step, performed by the at least one processor, of retrieving an answer according to the corresponding selected service,

wherein the service selecting step comprises an automatic service selecting step of searching lost content in a current query by using a service mapping rule base and complementing the lost content through searching a current user query history base,

wherein the automatic service selecting step comprises;

a lost content searching step of matching the semantically-analyzed query with a service mapping rule in the service mapping rule base, and extracting a service type to which the current query belongs and a lost parameter in the current query;

a latest query detecting step of searching a latest query contained the lost parameter in the current query from the current user query history base, and extracting the parameter value corresponding to the lost parameter in the current query; and

a query complementing step of adding the service type, the lost parameter and the parameter value into the semantically-analyzed query to form the complete query.

20. The method according to claim 19, wherein the automatic service selecting step further comprises a similar query detecting step of searching a history query that contains the lost parameter in the current query as a similar query from the current user query history base when there is no parameter value extracted by the latest query detecting step, and extracting the parameter value corresponding to the lost parameter in the current query from the similar query.

21. The method according to claim 20, wherein the similar query detecting step comprises a step of retrieving the history query meeting the following conditions as a similar query from the current user query history base:

a service type of the history query is identical to a service type to which the current query belongs; and

a query parameter of the history query contains a query parameter of the semantically-analyzed query.

22. The method according to claim 19, wherein the lost content searching step comprises a step of retrieving the service mapping rule meeting the following conditions as a matched service mapping rule from the service mapping rule base:

a requirement in the service mapping rule is identical to a requirement of the semantically-analyzed query; and

a service parameter in the service mapping rule contains a parameter semantically-analyzed query.

23. A method of selecting service based on an incomplete query, comprising:

a semantic analyzing step, performed by at least one processor, of analyzing an incomplete query from a user semantically;

a service selecting step, performed by the at least one processor, of complementing the incomplete query based on a semantically-analyzed query to complete the incomplete query, and selecting a corresponding selected service from among a plurality of services based on a complete query; and

a retrieving step, performed by the at least one processor, of retrieving an answer according to the corresponding selected service,

wherein the service selecting step comprises an automatic service selecting step of searching lost content in a current query by using a service mapping rule base and complementing the lost content through searching an other user query history base,

wherein the automatic service selecting step comprises;

a lost content searching step of matching the semantically-analyzed query with a service mapping rule in the service mapping rule base, and extracting a service type to which current query belongs and a lost parameter in the current query;

a similar query detecting step of searching a history query that contains the lost parameter in the current query as a similar query from the other user query history base, and extracting a parameter value corresponding to the lost parameter in the current query from the similar query; and

a query complementing step of adding the service type, the lost parameter and the parameter value into the semantically-analyzed query to form the complete query.

24. The method according to claim 23, wherein the lost content searching step comprises a step of retrieving the service mapping rule meeting the following conditions as matched service mapping rule from the service mapping rule base:

a requirement in the service mapping rule is identical to a requirement of the semantically-analyzed query; and

a service parameter in the service mapping rule contains a parameter of the semantically-analyzed query.

25. The method according to claim 23, wherein the similar query detecting step comprises a step of retrieving the history query meeting the following conditions as a similar query from the other user query history base:

a service type of the history query is identical to the service type to which the current query belongs; and

a query parameter of the history query contains a query parameter of the semantically-analyzed query.

26. A method of selecting service based on an incomplete query, comprising:

a semantic analyzing step, performed by at least one processor, of analyzing an incomplete query from a user semantically;

a service selecting step, performed by the at least one processor, of complementing the incomplete query based on a semantically-analyzed query to complete the incomplete query, and selecting a corresponding selected service from among a plurality of services based on a complete query; and

a retrieving step, performed by the at least one processor, of retrieving an answer according to the corresponding selected service,

wherein the service selecting step comprises an automatic service selecting step of searching lost content in a current query by using a service mapping rule base and complementing the lost content through searching a fact base,

wherein the automatic service selecting step comprises;

a lost content searching step of matching the semantically-analyzed query with a service mapping rule in the service mapping rule base, and extracting a service type to which the incomplete query belongs and a lost parameter in the incomplete query;

a fact matching step of matching the semantically-analyzed query with a fact in a fact base to find a matched fact, and extracting a fault value from the matched fact as a lost parameter value; and

a query complementing step of adding the service type, the lost parameter and the lost parameter value into the semantically-analyzed query to form the complete query.

27. The method according to claim 26, wherein the lost content searching step of retrieving the service mapping rule meeting the following conditions as a matched service mapping rule from the service mapping rule base:

a requirement in the service mapping rule is identical to a requirement of the semantically-analyzed query; and

a service parameter in the service mapping rule contains a parameter of the semantically-analyzed query.

28. The method according to claim 26, wherein the fact matching step of retrieving the fact meeting the following conditions as matched fact from the fact base:

a service type of the fact is identical to the service type to which the current query belongs; and

a lost parameter in the fact is identical to the lost parameter in the current query.

29. A query system, comprising:

a query receiver which receives a user query;

a semantic analyzing device which parses the user query and semantically analyzes the user query;

a service selecting device which complements an incomplete query based on the semantically-analyzed query to form the complete query, and selects a corresponding selected service from among a plurality of services based on a complete query;

a retrieving device which retrieves an answer according to the corresponding selected service;

an answer sender which sends the answer to a user; and

at least one processor coupled to the query receiver, the semantic analyzing device, the service selecting device, the retrieving device and the answer sender,

wherein the service selecting device comprises a semi-automatic service selecting section which searches lost content in the incomplete query by using a service mapping rule base and complements the lost content through an interacting with the user,

wherein the semi-automatic service selecting section comprises;

a lost content searching unit which matches the semantically-analyzed query with a service mapping rule in the service mapping rule base, and extracts a service type to which the incomplete query belongs and a lost parameter in the incomplete query;

a user interacting unit which prompts the user to input prompt information of a parameter value corresponding to the lost parameter and receives feedback information including the parameter value from the user;

a parameter value extracting unit which extracts the parameter value from the feedback information of the user; and

a query complementing unit which adds the service type, the lost parameter and the parameter value into the semantically-analyzed query to form the complete query.

30. A query method, comprising:

a query receiving step, performed by at least one processor, of receiving a user query;

a semantic analyzing step of parsing the user query and semantically analyzes the user query;

a service selecting step, performed by the at least one processor, of complementing an incomplete query based on the semantically-analyzed query to complete the incomplete query, and selecting a corresponding selected service from among a plurality of services based on a complete query;

a retrieving step, performed by the at least one processor, of retrieving an answer according to the corresponding selected service; and

an answer sending step, performed by the at least one processor, of sending the answer to a user,

wherein the service selecting step comprises a semi-automatic service selecting step of searching lost content in the incomplete query by using a service mapping rule base and complementing the lost content through an interacting with the user,

wherein the semi-automatic service selecting step comprises;

a lost content searching step of matching the semantically-analyzed query with a service mapping rule in the service mapping rule base, and extracting a service type to which the incomplete query belongs and a lost parameter in the incomplete query;

a user interacting step of prompting the user to input prompt information of a parameter value corresponding to the lost parameter and receiving feedback information including the parameter value from the user;

a parameter value extracting step of extracting the parameter value from the feedback information of the user; and

a query complementing step of adding the service type, the lost parameter and the parameter value into the semantically-analyzed query to form the complete query.

31. A query system, comprising:

a query receiver which receives a user query;

a semantic analyzing device which parses the user query and semantically analyzes the user query;

a determining device which determines whether the user query is a complete user query;

a first service selecting device which performs a process on the complete user query and selects a first selected service from among a plurality of services based on the complete user query;

a second service selecting device which complements an incomplete user query to complete the incomplete user query and selects a second selected service from among the plurality of services based on the completed incomplete user query;

a retrieving device which retrieves an answer according to the first selected service or the second selected service;

an answer sender which sends the answer to a user; and

at least one processor coupled to the query receiver, the semantic analyzing device, the determining device, the first service selecting device, the second service selecting device, the retrieving device and the answer sender,

wherein the second service selecting device comprises a semi-automatic service selecting section which searches lost content in the incomplete query by using a service mapping rule base and complements the lost content through an interacting with the user,

wherein the semi-automatic service selecting section comprises;

a lost content searching unit which matches the semantically-analyzed query with a service mapping rule in the service mapping rule base, and extracts a service type to which the incomplete query belongs and a lost parameter in the incomplete query;

a user interacting unit which prompts the user to input prompt information of a parameter value corresponding to the lost parameter and receives feedback information including the parameter value from the user;

a parameter value extracting unit which extracts the parameter value from the feedback information of the user; and

a query complementing unit which adds the service type, the lost parameter and the parameter value into the semantically-analyzed query to form the complete query.

32. The query system according to claim 31, wherein the determining device matches the semantically-analyzed user query with the service mapping rule in the service mapping rule base, if a matching is successful, the user query is sent to the first service selecting device;

and if the matching is not successful, the user query is sent to the second service selecting device.

33. The query system according to claim 32, wherein the determining device determines whether the matching is successful according to the following conditions:

a requirement of the service mapping rule is identical to a requirement of the semantically-analyzed query; and

a service parameter of the service mapping rule contains a parameter of the semantically-analyzed query.

34. A query method, comprising:

a query receiving step, performed by at least one processor, of receiving a user query;

a semantic analyzing step, performed by the at least one processor, of parsing the user query and semantically analyzes the user query;

a determining step, performed by the at least one processor, of determining whether the user query is a complete user query;

a first service selecting step, performed by the at least one processor, of performing a process on the complete user query and selecting a first selected service from among a plurality of services based on the complete user query;

a second service selecting step, performed by the at least one processor, of complementing an incomplete query to complete the incomplete user query and selecting a second selected service from among the plurality of services based on the completed incomplete user query;

a retrieving step, performed by the at least one processor, of retrieving an answer according to the first selected service or the second selected service; and

an answer sending step, performed by the at least one processor, of sending the answer to a user,

wherein the second service selecting step comprises a semi-automatic service selecting step of searching lost content in the incomplete query by using a service mapping rule base and complementing the lost content through an interacting with the user,

wherein the semi-automatic service selecting step comprises;

a lost content searching step of matching the semantically-analyzed query with a service mapping rule in the service mapping rule base, and extracting a service type to which the incomplete query belongs and a lost parameter in the incomplete query;

a user interacting step of prompting the user to input prompt information of a parameter value corresponding to the lost parameter and receiving feedback information including the parameter value from the user;

a parameter value extracting step of extracting the parameter value from the feedback information of the user; and

a query complementing step of adding the service type, the lost parameter and the parameter value into the semantically-analyzed query to form the complete query.

35. The query method according to claim 34, wherein the determining step comprises a step of matching the semantically-analyzed user query with the service mapping rule in the service mapping rule base, if the matching is successful, the user query is processed by the first service selecting step;

and if the matching is not successful, the user query is processed by the second service selecting step.

36. The query method according to claim 35, wherein the determining step determines whether the matching is successful according to the following conditions:

a requirement of the service mapping rule is identical to a requirement of the semantically-analyzed query; and

a service parameter of the service mapping rule contains a parameter of the semantically-analyzed query.