Indexing and searching entity-relationship data

• US 8,751,505 B2
• Filed: 03/11/2012
• Issued: 06/10/2014
• Est. Priority Date: 03/11/2012
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Carmel, David, Roitman, Haggai, Yogev, Sivan
• Primary Examiner(s): Beausoliel, Jr., Robert W.
• Assistant Examiner(s): ALLEN, NICHOLAS E
• Application Number:
  US13/417,248
• Publication Number:
  US 20130238631A1
• Time in Patent Office: 821 Days
• Field of Search: 705/7.14, 705/321, 707/694, 707/765, 707/781, 709/223, 709/224, 714/4.11, 714/6.3
• US Class Current: 707/742
• CPC Class Codes: G06F 16/288 Entity relationship models

Abstract

Method, system, and computer program product for indexing and searching entity-relationship data are provided. The method includes: defining a logical document model for entity-relationship data including: representing an entity as a document containing the entity'"'"'s searchable content and metadata; dually representing the entity as a document and as a category; and representing each relationship instance for the entity as a category set that contains categories of all participating entities in the relationship. The method also includes: translating entity-relationship data into the logical document model; and indexing the entity-relationship data of the populated logical document model as an inverted index. The method may include searching indexed entity-relationship data using a faceted search, wherein the categories are all categories required for supporting faceted navigation.

Claims

1. A computer-implemented method for indexing and searching entity-relationship data performed by a computerized device having a processor, comprising:

providing a database of entity-relationship data which includes identifiers of entities and indications of relationships between the entities, indicating which entities are related to each other;

representing the entities of the provided database as respective documents, each document containing content and metadata of a respective entity;

representing the entities of the provided database as respective categories;

representing each relationship instance, which indicates a relationship between a plurality of entities, in the database, as a category set that contains categories of all participating entities in the relationship; and

indexing the entity-relationship data of a populated logical document model as an inverted index, which indicates for each relationship attribute, the entities to which it relates, allowing lookup in the inverted index of results of queries based on relationship participation patterns.

2. The computer-implemented method of claim 1, wherein defining a logical document model for entity-relationship data includes representing a relationship between a plurality of entities as a category set that contains:

categories of all participating entities in the relationship; and

categories representing all attributes in the relationship; and

a category representing the relationship type.

3. The computer-implemented method of claim 1, wherein defining a logical document model for entity-relationship data includes representing a relationship between a plurality of entities as a document containing the relationship category set.

4. The computer-implemented method of claim 1, wherein defining a logical document model for entity-relationship data includes inclusion of a relationship category set in documents representing entities participating in the relationship.

5. The computer-implemented method of claim 1, wherein the entity-relationship data that can be indexed includes multiple entities, wherein each entity has none to multiple relationships, and a relationship is an association between two or more entities.

6. The computer-implemented method of claim 1, wherein the entity-relationship data that can be indexed includes:

entity data with multiple attributes of various data types, relationship data with arbitrary cardinality, and attributes on relationships denoting relationship context.

7. The computer-implemented method of claim 1, including:

searching indexed entity-relationship data using a faceted search, wherein the categories are all categories required for supporting faceted navigation.

8. The computer-implemented method of claim 1, wherein the logical document model defines fields for a document including one or more of the group of:

a key, a type, attributes, attribute names, free-text content of the entity or relationship for searching.

9. The computer-implemented method of claim 1, wherein the logical document model defines categories of attributes to be exposed as facets for searching.

10. The computer-implemented method of claim 1, wherein the logical document model includes for each field that represents a category, a payload that contains the identifiers of all category sets that include this category.

11. The computer-implemented method of claim 1, including:

obtaining a set of entities which comply with a query as a list of documents or as a list of categories with transition from one representation to the other performed using the fields in the logical document model.

12. The computer-implemented method of claim 1, including:

processing a free-text query by searching over a free-text content field encapsulating the searchable content of the entities.

13. The computer-implemented method of claim 1, including:

processing entity and attribute queries by retrieving matching entity and/or attributes from the inverted index.

14. The computer-implemented method of claim 1, including:

processing relationship queries by retrieving for each predicate in the query a list of categories which comply with the predicate, and then identifying category sets which contain for each predicate at least one category complying to the predicate.

15. The computer-implemented method of claim 1, including:

receiving a user query, by the processor;

providing by the processor a result set for the query;

receiving user input, via a graphical user interface, of related entities using a relationship facet, and

refining the user query to reflect the received user input, using structured query predicates.

16. The computer-implemented method of claim 1, including:

indexing and retrieval of documents with inheritance metadata, where the category representing an entity contains entity type inheritance information, and where type-specific processing of attributes is applied also for entities to which the type is a super-type.

17. A computer program product for indexing and searching entity-relationship data, the computer program product comprising:

a computer readable non-transitory storage medium having computer readable program code embodied therewith, the computer readable program code comprising;

computer readable program code configured to;

access a database of entity-relationship data which includes identifiers of entities and indications of relationships between the entities, indicating which entities are related to each other;

represent the entities of the provided database as respective documents, each document containing content and metadata of a respective entity;

represent the entities of the provided database as respective categories;

represent each relationship instance, which indicates a relationship between a plurality of entities, in the database, as a category set that contains categories of all participating entities in the relationship; and

index the entity-relationship data of a populated logical document model as an inverted index.

18. A system for indexing and searching entity-relationship data, comprising:

a processor configured to;

receive indication of a database of entity-relationship data which includes identifiers of entities and indications of relationships between the entities, indicating which entities are related to each other;

represent the entities of the indicated database as respective documents, each document containing content and metadata of a respective entity;

represent the entities of the provided database as respective categories;

represent each relationship instance, which indicates a relationship between a plurality of entities, in the database, as a category set that contains categories of all participating entities in the relationship; and

index the entity-relationship data of a populated logical document model as an inverted index.

19. The system of claim 18, including:

a searching system for obtaining a set of entities which comply with a query as a list of documents or as a list of categories with transition from one representation to the other performed using the fields in the logical document model.

20. The system of claim 18, including:

a free-text query component for processing a free-text query by searching over a free-text content field encapsulating the searchable content of the entities.

21. The system of claim 18, including:

an entity and attribute query component for processing entity and attribute queries by retrieving matching entity and/or attributes from the inverted index.

22. The system of claim 18, including:

a relationship query component for processing relationship queries.

23. The system of claim 18, including:

a graphical user interface for receiving user input of related entities using a relationship facet, refining user query to reflect choice using structured query predicates.

24. The system of claim 18, wherein the processor is configured to index documents with inheritance metadata.

25. A method of providing a service to a customer over a network for access control for indexing and searching entity-relationship data, the service comprising:

providing a database of entity-relationship data which includes identifiers of entities and indications of relationships between the entities, indicating which entities are related to each other;

representing the entities of the provided database as respective documents, each document containing content and metadata of a respective entity;

representing the entities of the provided database as respective categories;

representing each relationship instance, which indicates a relationship between a plurality of entities, in the database, as a category set that contains categories of all participating entities in the relationship; and

indexing the entity-relationship data of a populated logical document model as an inverted index.