Method, apparatus, and computer program product for categorizing multiple group-based communication messages

Patent metadata

Field	Value
Publication number	US-11115360-B2
Application number	US-201916523647-A
Country	US
Kind code	B2
Filing date	Jul 26, 2019
Priority date	Jul 30, 2018
Publication date	Sep 7, 2021
Grant date	Sep 7, 2021

Abstract

The example embodiments provided relate to a group-based communication system configured to categorize multiple communication messages in a workspace by determining one or more conversation primitive identifiers for each of the multiple communication messages. In some examples, the group-based communication system may group each of the multiple communication messages into one or more conversation segments based on the one or more conversation primitive identifiers.

First claim

The invention claimed is:

1. An apparatus for group-based communication message categorization, the apparatus comprising at least one processor and at least one non-transitory memory including program code, the at least one non-transitory memory and the program code configured to, with the at least one processor, cause the apparatus to:
receive a channel message corpus representing multiple communication messages of a group-based communication channel, the multiple communication messages of the channel message corpus sharing a group-based communication channel identifier corresponding to the group-based communication channel;
determine, using a conversation primitive identification engine coupled with the at least one processor, one or more conversation primitive identifiers for each of the multiple communication messages;
identify one or more communication message segments of the multiple communication messages that share at least one common conversation primitive identifier of the one or more conversation primitive identifiers; and
in response to identifying the one or more communication message segments that share at least one common conversation primitive identifier, group the one or more communication message segments that share the at least one common conversation primitive identifier into a conversation segment, wherein the conversation segment includes a topic associated with the conversation segment.

2. The apparatus of claim 1 , wherein the at least one non-transitory memory and the program code are further configured to, with the at least one processor, cause the apparatus to determine the one or more-conversation primitive identifiers for each of the multiple communication messages by:
identifying timestamp data associated with at least two of the multiple communication messages;
calculating a set of message send time proximities between the at least two of the multiple communication messages using the timestamp data;
comparing the message send time proximities to a pre-determined message send time threshold; and
in a circumstance where a selected subset of the message send time proximities satisfy the pre-determined message send time threshold, issuing the at least one common conversation primitive identifier to respective communication messages that are associated with the selected subset of the message send time proximities.

3. The apparatus of claim 1 , wherein the at least one non-transitory memory and the program code are further configured to, with the at least one processor, cause the apparatus to determine the one or more conversation primitive identifiers for each of the multiple communication messages by:
generating an order list indicating a chronological order in which at least two of the multiple communication messages were sent;
calculating a set of message send order proximities between the at least two of the multiple communication messages using the order list;
comparing the set of message send order proximities to a pre-determined message send order threshold; and
in a circumstance where a selected subset of the message send order proximities satisfy the pre-determined message send order threshold, issuing the at least one common conversation primitive identifier to respective communication messages that are associated with the selected subset of the message send order proximities.

4. The apparatus of claim 1 , wherein the at least one non-transitory memory and the program code are further configured to, with the at least one processor, cause the apparatus to determine the one or more conversation primitive identifiers for each of the multiple communication messages by:
identifying message sending user identifiers associated with at least two of the multiple communication messages;
calculating a set of sending user identifier proximities between the at least two of the multiple communication messages using the identified message sending user identifiers;
comparing the set of sending user identifier proximities to a pre-determined sending user identifier threshold; and
in a circumstance where a selected subset of the sending user identifier proximities satisfy the pre-determined sending user identifier threshold, issuing the at least one common conversation primitive identifier to respective communication messages that are associated with the selected subset of the sending user identifier proximities.

5. The apparatus of claim 1 , wherein the at least one non-transitory memory and the program code are further configured to, with the at least one processor, cause the apparatus to determine the one or more conversation primitive identifiers for each of the multiple communication messages by:
calculating message send time proximities between at least two of the multiple communication messages;
calculating message send order proximities between at least two of the multiple communication messages;
calculating sending user identifier proximities between at least two of the multiple communication messages;
calculating a set of overall message proximities using the message send time proximities, the message send order proximities, and the sending user identifier proximities;
comparing the set of overall message proximities to a pre-determined overall message threshold; and
in a circumstance where a selected subset of the overall message proximities satisfy the pre-determined overall message threshold, issuing the at least one common conversation primitive identifier to respective communication messages that are associated with the selected subset of the overall message proximities.

6. The apparatus of claim 1 , wherein the at least one non-transitory memory and the program code are further configured to, with the at least one processor, cause the apparatus to determine the one or more conversation primitive identifiers for each of the multiple communication messages by:
identifying a set of topics associated with at least two of the multiple communication messages, the set of topics including the topic associated with the conversation segment; and
determining the one or more conversation primitive identifiers for each of the multiple communication messages based at least in part on individual topics of the set of topics.

7. The apparatus of claim 6 , wherein identifying the set of topics comprises parsing the at least two of the multiple communication messages into the one or more communication message segments using a message parsing machine learning model.

8. The apparatus of claim 6 , wherein the at least one non-transitory memory and the program code are further configured to, with the at least one processor, cause the apparatus to:
provide the at least two of the multiple communication messages to a user message categorization interface;
provide the set of topics to the user message categorization interface;
receive user-generated suggestions associated with the set of topics; and
determine one or more updated conversation primitive identifiers for the at least two of the multiple communication messages based at least in part on the user-generated suggestions.

9. The apparatus of claim 8 , wherein the user-generated suggestions comprise one or more of: the topic associated with the conversation segment, a correctness indication, and a topic redetermination indication.

10. The apparatus of claim 1 , wherein the at least one non-transitory memory and the program code are further configured to, with the at least one processor, cause the apparatus to:
provide, via a group-based communication interface, an indication of a conversation, the indication based at least in part on the one or more communication message segments.

11. A method for gro

Assignees

• Slack Tech Inc

Inventors

• Talton Jerry
• Morrison Andrew S
• Singh Japinder
• Katzur Jon
• Gallagher John

Classifications

• H04L51/216 Primary

  Handling conversation history, e.g. grouping of messages in sessions or threads · CPC title

• H04L51/046

  Interoperability with other network applications or services · CPC title

• H04L12/185 Primary

  with management of multicast group membership · CPC title

• H04L51/16 Primary

  Electricity · mapped topic