Method and apparatus for updating a previously generated text

• US 9,336,193 B2
• Filed: 02/27/2015
• Issued: 05/10/2016
• Est. Priority Date: 08/30/2012
• Status: Active Grant

• Current Assignee: Arria Data2text Limited
• Original Assignee: Arria Data2text Limited
• Inventors: Logan, Alasdair James, Reiter, Ehud Baruch
• Primary Examiner(s): Patel, Manglesh M
• Application Number:
  US14/634,158
• Publication Number:
  US 20150169522A1
• Time in Patent Office: 438 Days
• Field of Search: 715/256, 715/234, 715/200, 715/264
• US Class Current: 1/1
• CPC Class Codes: G06F 16/353 into predefined classes
• CPC Class Codes: G06F 40/14 Tree-structured documents parsing G06F40/205; validation G06F40/226
• CPC Class Codes: G06F 40/166 Editing, e.g. inserting or deleting
• CPC Class Codes: G06F 40/20 Natural language analysis semantic analysis of natural language G06F40/30

Abstract

Methods, apparatuses, and computer program products are described herein that are configured to enable updating of an output text. In some example embodiments, a method is provided that comprises generating a new message for each updateable data element based on a predetermined indication. The method of this embodiment may also include determining a classification for each new message by comparing each new message with a corresponding message that describes the updateable data element. The method of this embodiment may also include generating an additional document plan tree that contains at least a portion of the new messages. The method of this embodiment may also include combining the additional document plan tree with an original document plan tree.

Claims

1. A method for transforming one or more input data streams that comprise data that is at least partially expressed in a non-linguistic format into a format that can be expressed linguistically in one or more output texts, that method comprising:

generating, using a natural language generation system that is configured to execute on a processor, an output text based on a set of messages and a relationship between the set of messages as defined by an original document plan tree, wherein a message represents a phrase or a simple sentence and are created in an instance in which a first input data stream comprises data that satisfies one or more message requirements;

generating a new message for one or more updateable data elements based on a predetermined indication and a second input data stream;

determining a classification for the one or more new messages by comparing the one or more new messages with a corresponding earlier message that describes a updateable data element of the one or more updateable data elements;

generating an additional document plan tree that contains at least a portion of the one or more new messages;

combining at least a portion of the additional document plan tree with at least a portion of the original document plan tree; and

generating, using the natural language generation system, an updated output text, wherein the updated output text is generated based on the combined original document plan tree and additional document plan tree, wherein the combined original document plan tree and additional document plan tree is configured to be realized into one or more natural language sentences.

2. A method according to claim 1, wherein the new message is configured to describe the updateable data element over a predetermined time interval.

3. A method according to claim 1, wherein the new message is configured to describe a most significant change in the updateable data element over a predetermined time interval.

4. A method according to claim 1, wherein a linguistic marker is added to the beginning of the additional document plan tree to signify that the new messages in the additional document plan tree are an update to the information about the updateable data elements in the original document plan tree.

5. A method according to claim 1, wherein the new message is generated for each of the one or more updateable data elements and a classification is determined for each of the one or more new messages.

6. A method according to claim 5, wherein the updateable data elements are identified in an instance in which they describe the data channel at a time of a triggering event.

7. A method according to claim 1, further comprising:

generating the additional document plan tree;

generating a document plan tree node in the additional document plan tree for each classification;

placing each new message within a document plan tree node, wherein the classification of the new message corresponds to the classification of the document plan tree node; and

ordering each new message within each document plan tree node in the additional document plan tree based on a location of a most recent corresponding message with the original document plan tree.

8. A method according to claim 7, further comprising:

generating an annotation for one or more of the document plan tree nodes in the additional document plan tree, wherein the annotation is configured to provide realization instructions.

9. A method according to claim 8, wherein the annotation is configured to override a default behavior.

10. A method according to claim 1, wherein the predetermined indication is a time period of at least one of a fixed interval, a regular interval until an acknowledgement is detected or based on a received request.

11. An apparatus configured to transform one or more input data streams that comprise data that is at least partially expressed in a non-linguistic format into a format that can be expressed linguistically in one or more output texts, the apparatus comprising:

at least one processor; and

at least one memory including computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least;

generate, via a natural language generation system that is configured to execute on a processor, an output based on a set of messages and a relationship between the set of messages as defined by an original document plan tree, wherein a message represents a phrase or a simple sentence and are created in an instance in which a first input data stream comprises data that satisfies one or more message requirements;

generate a new message for one or more updateable data elements based on a predetermined indication and a second input data stream;

determine a classification for the one or more new messages by comparing the one or more new messages with a corresponding earlier message that describes a updateable data element of the one or more updateable data elements;

generate an additional document plan tree that contains at least a portion of the one or more new messages; and

combine at least a portion of the additional document plan tree with at least a portion of the original document plan tree; and

generate, via the natural language generation system, an updated output text, wherein the updated output text is generated based on the combined original document plan tree and additional document plan tree, wherein the combined original document plan tree and additional document plan tree is configured to be realized into one or more natural language sentences.

12. An apparatus according to claim 11, wherein the new message is configured to describe the updateable data element over a predetermined time interval.

13. An apparatus according to claim 11, wherein the new message is configured to describe a most significant change in the updateable data element over a predetermined time interval.

14. An apparatus according to claim 11, wherein a linguistic marker is added to the beginning of the additional document plan tree to signify that the new messages in the additional document plan tree are an update to the information about the updateable data elements in the original document plan tree.

15. An apparatus according to claim 11, wherein the new message is generated for each of the one or more updateable data elements and a classification is determined for each of the one or more new messages.

16. An apparatus according to claim 15, wherein the updateable data elements are identified in an instance in which they describe the data channel at a time of a triggering event.

17. An apparatus according to claim 11, wherein the at least one memory including the computer program code is further configured to, with the at least one processor, cause the apparatus to:

generate the additional document plan tree;

generate a document plan tree node in the additional document plan tree for each classification;

place each of the new messages within a document plan tree node, wherein the classification of the new message corresponds to the classification of the document plan tree node; and

order each new message within each document plan tree node in the additional document plan tree based on a location of a most recent corresponding message with the original document plan tree.

18. An apparatus according to claim 17, wherein the at least one memory including the computer program code is further configured to, with the at least one processor, cause the apparatus to:

generate an annotation for one or more of the document plan tree nodes in the additional document plan tree, wherein the annotation is configured to provide realization instructions.

19. An apparatus according to claim 11, wherein the predetermined indication is a time period of at least one of a fixed interval, a regular interval until an acknowledgement is detected or based on a received request.

20. A computer program product configured to transform one or more input data streams that comprise data that is at least partially expressed in a non-linguistic format into a format that can be expressed linguistically in one or more output texts, the computer program product comprising:

at least one computer readable non-transitory memory medium having program code stored thereon, the program code which when executed by an apparatus cause the apparatus at least to;

generate, via a natural language generation system that is configured to execute on a processor, an output text based on a set of messages and a relationship between the set of messages as defined by an original document plan tree, wherein a message represents a phrase or a simple sentence and are created in an instance in which a first input data stream comprises data that satisfies one or more message requirements;

generate a new message for one or more updateable data elements based on a predetermined indication and a second input data stream;

determine a classification for the one or more new messages by comparing the one or more new messages with a corresponding earlier message that describes a updateable data element of the one or more updateable data elements;

generate an additional document plan tree that contains at least a portion of the new messages; and

combine at least a portion of the additional document plan tree with at least a portion of the original document plan tree; and

generate, via the natural language generation system, an updated output text, wherein the updated output text is generated based on the combined original document plan tree and additional document plan tree, wherein the combined original document plan tree and additional document plan tree is configured to be realized into one or more natural language sentences.