Systems and methods for automated content generation

• US 8,457,948 B2
• Filed: 05/13/2010
• Issued: 06/04/2013
• Est. Priority Date: 05/13/2010
• Status: Active Grant

• Current Assignee: Expedia Inc. (Expedia Group, Inc.)
• Original Assignee: Expedia Inc. (Expedia Group, Inc.)
• Inventors: Waksberg, Rene, Hsu, Donny, Bradley, Patrick
• Primary Examiner(s): Desir, Pierre-Louis
• Assistant Examiner(s): KOVACEK, DAVID M
• Application Number:
  US12/779,902
• Publication Number:
  US 20110282649A1
• Time in Patent Office: 1,118 Days
• Field of Search: 704 1- 8, 704 9- 10, 707/705, 707748-751, 707828-831, 707E17001-E17143
• US Class Current: 704/8
• CPC Class Codes: G06F 16/972 Access to data in other repository systems, e.g. legacy data or dynamic Web page generation
• CPC Class Codes: G06F 40/186 Templates

Abstract

A content generation service is described that generates content for electronic documents in different languages based upon templates. The templates may include paragraph templates composed of sentence types including sentence templates. The sentence templates may further include variables having corresponding attributes. Each of the paragraph templates, sentence templates, and attributes may be hierarchically organized. The content generation service may obtain data describing an item of interest, such as a travel item. The obtained data may further specify a document language, section and paragraph for which content is to be generated. Content is generated for variables in hierarchical order, with higher ranked paragraphs considered first. Within the highest ranked paragraph, a sentence type is selected and the variables within the highest ranked sentence template of the sentence type are considered. Sentence templates which contain attributes matched by the data obtained for the item of interest are added to the generated document.

Claims

1. A computer-implemented method of electronically generating a document describing a property in different languages, the method comprising:

under control of one or more configured computer systems;

obtaining first data, wherein the first data comprises a set of attributes associated with the property;

obtaining a content template for generation of content describing the property, the content template comprising;

a plurality of paragraph templates, wherein each of the plurality of paragraph templates comprises one or more sentence types, and wherein the plurality of paragraph templates are ranked at least according to the desirability of using each paragraph template to generate content describing the property;

a plurality of sentence templates, wherein each of the plurality of sentence templates corresponds to at least one of the one or more sentence types, wherein each of the plurality of sentence templates comprises one or more variables, and wherein the plurality of sentence templates are ranked at least according to the desirability of using each sentence template to generate content describing the property; and

wherein the plurality of paragraph templates and the plurality of sentence templates are ranked independent of the set of attributes associated with the property;

obtaining second data, the second data specifying the language of the template;

identifying the highest ranked paragraph template of the plurality of paragraph templates based at least in part on the second data;

for each sentence type of the identified paragraph template;

identifying the highest ranked sentence template of the plurality of sentence templates that corresponds to the sentence type; and

determining that each of the variables of the identified sentence template correspond to at least one of the set of attributes of the first data; and

generating a document based at least in part on processing the identified paragraph template with at least a sub-set of the set of attributes.

2. The computer-implemented method of claim 1, wherein obtaining a content template for generation of content describing the property comprises:

identifying a language specified within the obtained second data; and

retrieving a content template for the property which is written in the identified language.

3. The computer-implemented method of claim 1, wherein the obtained first data comprises one or more of a property name, property amenities, a property rating, a property location, and points of interest associated with the property.

4. The computer-implemented method of claim 1, wherein the generated document does not include unsatisfied sentence templates, wherein an unsatisfied sentence template has at least one variable that does not correspond to an attribute within the set of attributes of the first data.

5. The computer-implemented method of claim 1, wherein one or more sentence types within the plurality of paragraph templates are designated as being required.

6. The computer-implemented method of claim 5, wherein the generated document does not include unsatisfied paragraph templates, wherein an unsatisfied paragraph template includes at least one sentence type corresponding to an unsatisfied sentence template, wherein an unsatisfied sentence template has at least one variable that does not correspond to an attribute within the set of attributes of the first data.

7. A system for automatically generating content for an electronic document, the system comprising:

an electronic data store that stores one or more templates for the electronic documents, wherein the one or more templates comprise;

a plurality of paragraph templates for a document paragraph, each paragraph template comprising one or more sentence types, wherein the plurality of paragraph templates are ranked at least according to a desirability of using each paragraph template to generate content describing items of interest,

a plurality of sentence templates, each sentence template corresponding to at least one of the one or more sentence types and comprising one or more variables, wherein the plurality of sentence templates are ranked at least according to a desirability of using each sentence template to generate content describing, items of interest; and

a content generation service implemented by one or more processors in communication with the electronic data store, the content generation service operative to;

obtain data regarding an item of interest to be described in the electronic document, the data comprising a set of attributes associated with the item of interest;

select a paragraph template which is to be employed for content generation from the data store based at least in part on the set of attributes and on rankings of the paragraph templates, wherein the ranking of the plurality of paragraph templates is independent of the set of attributes;

for each sentence type of the selected paragraph template, identify a sentence template of the plurality of sentence templates that corresponds to said sentence type based at least in part on a rank of the identified sentence template and on the set of attributes, wherein each of the variables of the identified sentence template correspond to at least one attribute of the set of attributes, and wherein the ranking of the plurality of sentence templates is independent of the set of attributes; and

generate the electronic document based at least in part on processing the selected paragraph template with at least a portion of the set of attributes.

8. The system of claim 7, wherein the item of interest comprises at least one of flights, lodging properties, ground transportation, and cruises.

9. The system of claim 7, wherein data regarding the item of interest comprises one or more of a property name, property amenities, a property rating, a property location, and points of interest associated with the property.

10. The system of claim 7, wherein the set of input attributes comprise at least one attribute type and at least one corresponding attribute value.

11. The system of claim 10, wherein the variables of the identified sentence template include an indicator specifying whether the sentence template variables are replaced by an attribute type or attribute value.

12. The system of claim 7, wherein selecting a paragraph template comprises:

obtaining additional data identifying a desired language of the electronic document; and

retrieving a paragraph template for the property which is written in the desired language.

13. The system of claim 7, wherein selecting a paragraph template comprises selecting the highest ranked paragraph template that may be used to generate a paragraph describing the item of interest based at least in part on the set of attributes.

14. The system of claim 7, wherein the generated electronic document does not include unsatisfied sentence templates, wherein an unsatisfied sentence template has at least one variable that does not correspond to an attribute within the set of attributes.

15. The system of claim 7, wherein one or more sentence types within the selected paragraph template are designated as being required.

16. The system of claim 15, wherein the generated document does not include unsatisfied paragraph templates wherein an unsatisfied paragraph template includes at least one sentence type corresponding to an unsatisfied sentence template, wherein an unsatisfied sentence template has at least one variable that does not correspond to an attribute within the set of attributes.

17. A computer-implemented method of electronically generating content for a document describing an item of interest, the method comprising:

under control of one or more configured computer systems;

receiving a request to generate content for the item of interest;

obtaining data regarding the item of interest, the data including one or more attributes describing the item of interest for use in content generation and one or more languages for content generation;

identifying a template based at least in part upon the one or more languages for content generation and on the item of interest, wherein the template comprises;

a plurality of paragraph templates, each paragraph template comprising one or more sentence types, and

a plurality of sentence templates, each sentence template corresponding to at least one of the one or more sentence types and including one or more variables,

wherein the plurality of paragraph templates and the plurality of sentence templates are ranked at least according to a desirability of using each paragraph template and sentence template to generate content describing the item of interest, and wherein the ranking of the plurality of paragraph templates and the plurality of sentence templates is independent of the obtained data regarding the item of interest;

identifying a paragraph template of the plurality of paragraph templates based at least in part on the one or more attributes and on a ranking of the identified paragraph template;

for each sentence type of the selected paragraph template, identifying a sentence template of the plurality of sentence templates that corresponds to said sentence type based at least in part on a rank of the identified sentence template and on the set of attributes, wherein each of the variables of the identified sentence template correspond to at least one attribute of the set of attributes; and

generating a document, based at least in part on processing the identified paragraph template with at least a portion of the one or more attributes.

18. The computer-implemented method of claim 17, wherein the item of interest comprises at least one of flights, lodging properties, ground transportation, and cruises.

19. The computer-implemented method of claim 17, wherein data regarding the item of interest comprises one or more of a property name, property amenities, a property rating, a property location, and points of interest associated with the property.

20. The computer-implemented method of claim 17, wherein the attributes of the obtained data comprise at least one attribute type and at least one corresponding attribute value.

21. The computer-implemented method of claim 20, wherein the plurality of sentence templates further include an indicator specifying whether each sentence template variables is replaced by an attribute type or attribute value.

22. The computer-implemented method of claim 17, wherein the generated document does not include unsatisfied sentence templates, wherein an unsatisfied sentence template has at least one variable that does not correspond to an attribute within the one or more attributes of the obtained data.

23. The computer-implemented method of claim 17, wherein one or more sentence types of the plurality of paragraph templates are designated as being required.

24. The computer-implemented method of claim 23, wherein the generated document does not include unsatisfied paragraph templates, wherein an unsatisfied paragraph template includes at least one sentence type corresponding to an unsatisfied sentence template, wherein an unsatisfied sentence template has at least one variable that does not correspond to an attribute within the one or more attributes of the obtained data.

25. A non-transitory computer-readable medium having instructions encoded thereon, wherein the instructions, when executed by a computing apparatus, cause the computing apparatus to:

control a template component operative to;

obtaining a set of attributes associated with an item of interest;

select a paragraph template of a plurality of paragraph templates for a paragraph describing an item of interest, wherein each paragraph template comprises one or more sentence types, wherein each sentence type corresponds to at least one of a plurality of sentence templates, wherein each sentence template comprises one or more variables, and wherein the plurality of paragraph templates and the plurality of sentence templates are ranked independent of the set of attributes and at least according to a desirability of using each paragraph template and sentence template to generate content describing the item of interest;

for each sentence type of the selected paragraph template, identify a sentence template of the plurality of sentence templates that corresponds to said sentence type based at least in part on a rank of the identified sentence template and on the set of attributes, wherein each of the variables of the identified sentence template correspond to at least one attribute of the set of attributes;

and

generate an electronic document based at least in part on processing the selected paragraph template with at least a portion of the set of attributes.

26. The non-transitory computer-readable medium of claim 25, wherein the item of interest comprises at least one of flights, lodging properties, ground transportation, and cruises.

27. The non-transitory computer-readable medium of claim 25, wherein set of attributes comprises one or more of a property name, property amenities, a property rating, a property location, and points of interest associated with the property.

28. The non-transitory computer-readable medium of claim 25, wherein the set of attributes comprises at least one attribute type and at least one corresponding attribute value.

29. The non-transitory computer-readable medium of claim 28, wherein the plurality of sentence templates include an indicator specifying whether the sentence template variables are replaced by an attribute type or an attribute value.

30. The non-transitory computer-readable medium of claim 25, wherein selecting a paragraph template comprises:

obtaining additional data identifying a desired language for the paragraph describing the item of interest; and

retrieving a paragraph template from the plurality of paragraph templates which is written in the desired language.

31. The non-transitory computer-readable medium of claim 25, wherein selecting the paragraph template comprises selecting the highest ranked paragraph template of the plurality of paragraph templates that may be used to generate a paragraph describing the item of interest based at least in part on the set of attributes.

32. The non-transitory computer-readable medium of claim 25, wherein the generated document does not include unsatisfied sentence templates, wherein an unsatisfied sentence template has at least one variable that does not correspond to an attribute within the set of attributes.

33. The non-transitory computer-readable medium of claim 25, wherein one or more sentence types of the plurality of paragraph templates are designated as being required.

34. The non-transitory computer-readable medium of claim 33, wherein the generated document does not include unsatisfied paragraph templates, wherein an unsatisfied paragraph template includes at least one sentence type corresponding to an unsatisfied sentence template, wherein an unsatisfied sentence template has at least one variable that does not correspond to an attribute within the set of attributes.