Context-aware aggregation of text-based messages

• US 9,705,832 B2
• Filed: 08/27/2014
• Issued: 07/11/2017
• Est. Priority Date: 08/27/2014
• Status: Active Grant

• Current Assignee: Lenovo PC International Limited (Lenovo Group Ltd.)
• Original Assignee: Lenovo Singapore Pte Limited (Lenovo Group Ltd.)
• Inventors: VanBlon, Russell Speight, Mese, John Carl, Waltermann, Rod D., Peterson, Nathan J.
• Primary Examiner(s): Vu, Viet
• Assistant Examiner(s): EDWARDS, JAMES A
• Application Number:
  US14/470,710
• Publication Number:
  US 20160065519A1
• Time in Patent Office: 1,049 Days
• Field of Search: None
• CPC Class Codes: H04L 51/216 Handling conversation history, e.g. grouping of messages in sessions or threads
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46

Abstract

For automatically aggregating text-based messages together based on context, an apparatus for automatically identifying a topic of a reply message and associating the reply message with a conversation thread based on the topic is disclosed. The apparatus may include a processor and a memory that stores code executable by the processor, including code that identifies a messaging conversation between two or more messaging devices, code that determines a topic of a reply message in the messaging conversation, and code that associates the message with a conversation thread based on the topic. In one embodiment, the apparatus examines active applications on a messaging device composing the message and determines the topic based on the active applications. In one embodiment, the apparatus identifies an unanswered question and determines whether the reply message answers the unanswered question. A method and computer program product also perform the functions of the apparatus.

Claims

1. An apparatus comprising:

a processor;

a memory that stores code executable by the processor, the code comprising;

code that identifies a conversation supporting a plurality of conversation threads, the conversation comprising a series of messages among two or more devices;

code that tracks eye movement;

code that determines a viewing location based on the eye movement;

code that determines a topic of a message in the conversation, wherein determining the topic is based on the onscreen content;

code that determines whether the topic matches an existing conversation thread of the conversation;

code that associates the message with the existing conversation thread in response to the topic matching;

code that creates a new conversation thread in the conversation in response to the topic not matching; and

code that visually distinguishes the conversation threads within the conversation.

2. The apparatus of claim 1, further comprising code that detects a reply message and automatically adds the reply message to a conversation thread based on context of the reply message.

3. The apparatus of claim 1, further comprising code that examines active applications on a device composing the message, wherein determining the topic is based on the active applications.

4. The apparatus of claim 1, further comprising code that associates the message with a particular conversation thread of the conversation based on user input.

5. The apparatus of claim 1, further comprising:

code that creates a data structure associating keywords with each conversation thread of the conversation; and

code that matches content in the message with keywords of the data structure, wherein determining the topic of the message is based on content in the message.

6. The apparatus of claim 1, wherein the code that determines the topic of a message comprises identifying an unanswered question and determining a likelihood that the message answers the unanswered question.

7. The apparatus of claim 1, wherein the code that determines the topic of a message comprises automatically determining the topic of a latest message in response to an action selected from the group comprising sending, receiving, and composing the latest message.

8. The apparatus of claim 1, wherein the apparatus comprises a device selected from the group consisting of:

a receiving device, a sending device, and a messaging server.

9. A method comprising:

identifying, by use of a processor, a conversation supporting a plurality of message groupings, the conversation comprising a series of messages among two or more devices;

tracking eye movement;

determining a viewing location based on the eye movement;

identifying a topic of a message in the conversation, wherein determining the topic is based on the onscreen content;

determining whether the identified topic belongs to a message grouping of the conversation;

associating the message with the message grouping in response to determining that the message belongs to the message grouping;

creating a new message grouping in the conversation in response to determining that the message does not belong to an existing message grouping of the conversation; and

visually distinguishing the conversation threads within the conversation.

10. The method of claim 9, wherein the message grouping is a conversation thread, the method further comprising:

associating the message with the new conversation thread in response to determining that the message does not belong to any conversation threads.

11. The method of claim 9, further comprising receiving audio data of a device composing the message, wherein identifying a topic of the message comprises identifying the topic based on the received audio data.

12. The method of claim 9, further comprising creating a context stack for the conversation, the context stack comprising context associated with each message of the conversation, wherein determining whether the identified topic belongs to a message grouping comprises searching the context stack for context matching the identified topic.

13. The method of claim 9, further comprising associating keywords with each message in the conversation, wherein determining whether the identified topic belongs to a message grouping comprises comparing the identified topic with the keywords.

14. The method of claim 9, wherein identifying a topic of the message comprises identifying a cursor location of a device composing the message.

15. The method of claim 9, wherein identifying a topic of the message comprises performing a natural language analysis of context of the message.

16. A program product comprising a computer readable storage medium that stores code executable by a processor, the executable code comprising code to perform:

identifying a latest message in conversation supporting a plurality of conversation threads, the conversation comprising a series of messages among two or more messaging devices;

tracking eye movement;

determining a viewing location based on the eye movement;

determining a topic of the latest message, wherein determining the topic is based on the onscreen content;

determining whether the topic matches an existing conversation thread of the conversation;

associating the latest message with the existing thread of the conversation in response to the topic matching;

creating a new conversation thread for the latest message in response to the topic not matching; and

visually distinguishing the conversation threads within the conversation.

17. The program product of claim 16, the code further comprising identifying keywords for each of a plurality of threads associated with the conversation, wherein determining the topic comprises searching the message for keywords associated with a thread.

18. The program product of claim 16, wherein associating the latest message with a thread comprises:

calculating a likelihood that topic of the latest message matches a theme of each thread associated with the conversation;

identifying a particular thread having a highest calculated likelihood;

comparing the highest calculated likelihood to a threshold;

associating the latest message with the particular thread in response to the highest calculated likelihood exceeding the threshold; and

associating the latest message with a new thread in response to the highest calculated likelihood not exceeding the threshold.