Mixed initiative semantic search

• US 20080319947A1
• Filed: 06/25/2007
• Published: 12/25/2008
• Est. Priority Date: 06/25/2007
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP AG (SAP SE)
• Inventors: Latzina, Markus, Soltani, Anoshirwan
• Granted Patent: US 7,987,176 B2
• Time in Patent Office: Days
• US Class Current: 707/3
• CPC Class Codes: G06F 16/334 Query execution G06F16/335 takes precedence
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

In one embodiment, a method is illustrated including receiving a search query, parsing the search query to identify first and second search terms, determining a relationship between the first and second search terms based on their respective domain assignments, conducting a search based on the respective domain assignments, and displaying a result of the search based on the respective domain assignments, and the relationship between the first and second search terms, as at least one rephrasing of the search query.

Claims

1. A method comprising:

receiving a search query;

parsing the search query to identify first and second search terms;

determining a relationship between the first and second search terms based on their respective domain assignments;

conducting a search based on the respective domain assignments; and

displaying a result of the search based on the respective domain assignments, and the relationship between the first and second search terms, as at least one rephrasing of the search query.

2. The method of claim 1, further comprising assigning one of the relationship between the first and second search terms to a data store as an inter-domain relationship.

3. The method of claim 1, further comprising assigning the relationship between the first and second search terms to a data store as an intra-domain relationship.

4. The method of claim 1, further comprising assigning the first and second search terms to a domain.

5. The method of claim 1, further comprising assigning the first and second search terms to different domains.

6. The method of claim 1, wherein the relationship between the first and second search terms based on their respective domain assignments is at least one of a default relationship, a manually assigned relationship, or a probability-based relationship.

7. A method comprising:

parsing a search query into a constituent term;

retrieving an attribute of a domain based upon a correspondence between the constituent term and the attribute of the domain;

performing a database operation on the attribute of the domain and a second attribute of the domain, the database operation to generate a set of related data;

extracting the attribute of the domain from the set of related data;

relating the attribute of the domain to a semantic phrase, the semantic phrase representing a semantic relationship between the attribute of the domain and the second attribute of the domain; and

generating a semantic query using the semantic phrase and the attribute of the domain.

8. The method of claim 7, wherein the database operation includes at least one of a join, an inner join, an outer join, a Cartesian product, a difference, a union, or an intersection.

9. The method of claim 7, further comprising selecting a semantic phrase using a default relation, a probability value, or an Artificial Intelligence (AI) algorithm.

10. The method of claim 7, further comprising generating alternative semantic queries, the alternative semantic queries including the first member of the domain and a different semantic phrase.

11. The method of claim 7, wherein the semantic query is a Natural Language (NL) statement.

12. The method of claim 7, further comprising displaying the semantic query.

13. The method of claim 7, wherein the attribute and the second attribute are members of different domains.

14. A method comprising:

parsing training data to identify a constituent term; and

assigning the constituent term to a domain, the assigning being at least one of a default assignment, a percentile based assignment, or a manual assignment.

15. The method of claim 14, further comprising:

assigning the constituent terms to a domain as a node; and

linking the node to another node using a semantic relationship to generate a semantic map.

16. The method of claim 15, wherein the semantic map includes at least one of nodes linked within the same domain, or nodes linked within different domains.

17. A method comprising:

identifying a plurality of domains, each domain having one or more members; and

relating each of the one or more members using semantic relations organized into a hierarchy, the hierarchy based upon that semantic relation which is most commonly used relative to other semantic relations.

18. A computer system comprising:

a parser to parse a search query into one or more constituent terms;

a retrieving engine to retrieve a member of a domain based upon a correspondence between the one or more constituent terms and the member of the domain, the member of the domain being an attribute in a database table;

a relation engine to relate the member of the domain to a semantic phrase, the semantic phrase representing a semantic relationship between the member of the domain and another term; and

a generator to generate a semantic query using the semantic phrase and the member of the domain.

19. The computer system of claim 18, further comprising a selector to select the semantic phrase using a default relation, a probability value, or an Artificial Intelligence (AI) algorithm.

20. The computer system of claim 18, wherein the generator is used to generate alternative semantic queries, the alternative semantic queries including the member of the domain and a different semantic phrase.

21. The computer system of claim 18, wherein the semantic query is a Natural Language (NL) statement.

22. The computer system of claim 18, further comprising a display to display the semantic query.

23. In a computer system having a User Interface (UI) including a display and a selection device, a method of providing and selecting from a pane on the display, the method comprising:

retrieving data to be displayed in the pane, the data comprising at least one of a semantic query data, or search result data;

displaying the semantic query data, or search result data in the pane;

receiving a selection input signal indicative of the selection device identifying a selected position within the pane; and

in response to the selection input signal, presenting at least one of the semantic query data, or search result data.

24. The computer system having a UI of claim 23, further comprising displaying a search query in an input area.

25. The computer system having a UI of claim 23, further comprising displaying a label relating to a domain as a domain label.

26. The computer system having a UI of claim 23, further comprising displaying a semantic query in a best hits header area.

27. The computer system having a UI of claim 23, further comprising displaying the search result data in a best hits area.

28. The computer system having a UI of claim 23, further comprising displaying the semantic query data in a relations area.

29. The computer system having a UI of claim 23, further comprising displaying search results in an additional results area.

30. The computer system having a UI of claim 23, further comprising displaying search results in a related actions area.

31. A machine-readable medium comprising instructions, which when implemented by one or more machines cause the one or more machines to perform the following operations:

receiving a search query;

parsing the search query to identify one or more search terms;

determining at least one relationship between the one or more search terms based on their respective domain assignments;

conducting a search based on a domain relationship; and

displaying a result of the search related to the domain relationship as at least one rephrasing of the search query.