Back to results
US 11,960,546 B2Granted
Summarization of ambiguous strings
AssigneeInternational Business Machines Corporation
InventorsAdam John Gibbons, Seumas McLean Goddard, Shivani Joshi, James Kirk, Daphne Coates
Life of this patent
Estimating the term…
Abstract
A processor may analyze one or more unread messages. The one or more unread messages may be from one or more respective conversations. The processor may generate, from the analyzing, an ambiguous strings model. The processor may summarize, via utilization of the ambiguous strings model, the one or more unread messages. The processor may genericize the ambiguous strings model.
Claims20
A system for summarization of ambiguous strings, the system comprising:
- a memory;
- and
- a processor in communication with the memory, the processor being configured to perform operations comprising:
- analyzing two or more unread messages received on a cloud computing node, wherein the two or more unread messages are from one or more respective digital conversations, wherein analyzing the two or more unread messages comprises:
- separating and grouping, by a topic modeling component, topics of conversation in the two or more unread messages based on an identified content;
- identifying, by a sentiment analysis, an underlying intent of the two or more unread messages associated with the one or more respective digital conversations;
- and
- utilizing a linguistic analysis on the two or more unread messages from the one or more respective conversations to generate a 1:1 mapping of words/definition;
- generating, from the analyzing, an ambiguous strings model;
- summarizing, via utilization of the ambiguous strings model, the two or more unread messages from the one or more respective digital conversations by creating a single short message accounting for ambiguous text, wherein summarizing the two or more unread messages includes:
- utilizing the 1:1 mapping of words/definition to replace nonconventional characters in the two or more unread messages with a single definition associated with the nonconventional characters in the two or more unread messages;
- and
- genericizing the ambiguous strings model.
10 more elements
A computer-implemented method for summarization of ambiguous strings, the method comprising:
- analyzing, by a processor, two or more unread messages received on a cloud computing node, wherein the two or more unread messages are from one or more respective digital conversations, wherein analyzing the two or more unread messages comprises:
- separating and grouping, by a topic modeling component, topics of conversation in the two or more unread messages based on an identified content;
- identifying, by a sentiment analysis, an underlying intent of the two or more unread messages associated with the one or more respective digital conversations;
- and
- utilizing a linguistic analysis on the two or more unread messages from the one or more respective conversations to generate a 1:1 mapping of words/definition;
- generating, from the analyzing, an ambiguous strings model;
- summarizing, via utilization of the ambiguous strings model, the two or more unread messages from the one or more respective digital conversations by creating a single short message accounting for ambiguous text, wherein summarizing the two or more unread messages includes:
- utilizing the 1:1 mapping of words/definition to replace nonconventional characters in the two or more unread messages with a single definition associated with the nonconventional characters in the two or more unread messages;
- and
- genericizing the ambiguous strings model.
7 more elements
A computer program product for summarization of ambiguous strings comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform operations, the operations comprising:
- analyzing, by a processor, two or more unread messages received on a cloud computing node, wherein the two or more unread messages are from one or more respective digital conversations, wherein analyzing the two or more unread messages comprises:
- separating and grouping, by a topic modeling component, topics of conversation in the two or more unread messages based on an identified content;
- identifying, by a sentiment analysis, an underlying intent of the two or more unread messages associated with the one or more respective digital conversations;
- and
- utilizing a linguistic analysis on the two or more unread messages from the one or more respective conversations to generate a 1:1 mapping of words/definition;
- generating, from the analyzing, an ambiguous strings model;
- summarizing, via utilization of the ambiguous strings model, the two or more unread messages from the one or more respective digital conversations by creating a single short message accounting for ambiguous text, wherein summarizing the two or more unread messages includes:
- utilizing the 1:1 mapping of words/definition to replace nonconventional characters in the two or more unread messages with a single definition associated with the nonconventional characters in the two or more unread messages;
- and
- genericizing the ambiguous strings model.
7 more elements
Dependencies are inferred from the claim text.