Intelligent messaging assistant based on content understanding and relevance

• US 10,382,379 B1
• Filed: 06/15/2016
• Issued: 08/13/2019
• Est. Priority Date: 06/15/2015
• Status: Active Grant

• Current Assignee: Linfo IP LLC (Pueblo Nuevo LLC)
• Original Assignee: Guangsheng Zhang
• Inventors: Zhang, Guangsheng
• Primary Examiner(s): Vu, Viet D
• Application Number:
  US15/182,598
• Time in Patent Office: 1,154 Days
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: H04L 51/48 Message addressing, e.g. address format or anonymous messages, aliases

Abstract

System and methods for assisting a user to communicate a message to a recipient without requiring the user to provide a communication address of the recipient, especially when the address of a candidate recipient is unknown to the user, or the user has never communicated with the receiving party before, such as for sending a message to a company or store regarding a product/service, or sending an opinion/comment to a government agency or legislature, or to a social media site, etc. The methods are based on understanding the key information in an input content, such as a textual/audio/video message or an image, and detects objects, topics or domains of the input content; determining a relevance measure between the content in the message and a communication address of a potential recipient, and making an address suggestion or automatically send the message to the recipient, without requiring the user to spend much time and efforts in searching for the communication addresses of an intended recipient or learning how to use related communication tools.

Claims

1. A computer-implemented method for assisting communications, comprising:

receiving an input content from a user, wherein the input content contains two or more terms each comprising a word or a phrase, wherein the input content is received without also receiving a recipient address from the user to send the input content to;

identifying at least two terms in the text content, with one of the at least two terms identified as representing a topic, and the other of the at least two terms identified as representing a comment or opinion on a topic;

identifying a first term in the input content, and determining that the first term represents a comment or opinion on a first topic;

receiving, from a data source, a first communication address, wherein the first communication address is associated with a first entity, wherein the first entity includes a person or an organization that does not have a known personal or organizational relation with the user, wherein the data source is compiled by incorporating a textual description about an attribute associated with a business or activity conducted by the first entity, wherein the first communication address is received from a source other than the user'"'"'s collection of addresses, and the first communication address does not exist in the user'"'"'s collection of addresses and does not exist in the user'"'"'s organization'"'"'s collection of addresses, wherein the user'"'"'s collection of addresses is defined as a dataset containing addresses or contact information that is entered by the user or entered with the user'"'"'s permission, or containing the addresses or entities that the user has previously communicated with;

identifying a second term associated with the first communication address;

determining that the second term is relevant to the first topic;

identifying a relation between the first term and the first communication address based on the first term and the second term;

determining, based on the relation, the first communication address or the first entity as a candidate for the user to send the input content to, wherein the user does not know which electronic communication address or which entity the input content should be sent to; and

outputting the first communication address or its representation for use as a suggestion, or providing or displaying the first communication address or its representation to the user, wherein the representation includes a name or description of the first communication address or a name or description of the first entity.

2. The method of claim 1, further comprising:

providing an option to send the input content to the first communication address either using the user'"'"'s own identity or a proxy identity on behalf of the user.

3. The method of claim 1, further comprising:

determining a relevance measure between the input content and the first communication address or the first entity based on the relation;

when two or more communication addresses are displayed, ranking the two or more communication addresses or their representations based on the respective relevance measure.

4. The method of claim 1, wherein the second term is a term in a name or description associated with the first communication address, wherein the relation is identified by matching the first term with the second term.

5. The method of claim 1, wherein the second term is a term in a name or description associated with the first entity, wherein the relation is identified by matching the first term with the second term.

6. The method of claim 1, wherein the first term represents a first topic or domain and the second term represents a second topic or domain associated with the first entity, wherein the relation is identified by determining a relevance measure between the first topic or domain and the second topic or domain.

7. The method of claim 6, wherein the first term is not a name of the first entity.

8. The method of claim 6, wherein the input content does not contain a name of the first entity.

9. The method of claim 6, wherein the first term does not match the second term.

10. The method of claim 1, wherein the first communication address is in the form of a character string, and the second term is a sub-string of the first communication address, wherein the relation is identified by matching the first term with the second term.

11. A computer-implemented method for assisting communications, comprising:

receiving an input content from a user, wherein the input content contains one or more terms each comprising a word or a phrase, wherein the input content is received without also receiving a recipient address from the user to send the input content to;

identifying a first term in the input content;

receiving a first communication address from a data source, wherein the first communication address is associated with a first entity, wherein the first entity includes a person or an organization, wherein the first communication address is selected from a collection of addresses previously used by the user;

identifying a second term associated with the first communication address;

identifying a relation between the first term and the first communication address based on the first term and the second term;

determining, based on the relation, the first communication address as a potential candidate for the user to send the input content to, wherein the user does not know at least one electronic communication address or at least one entity the input content should be sent to; and

outputting the first communication address or its representation for use as a suggestion of a recipient to receive the input content, or providing or displaying the first communication address or its representation to the user, wherein the representation includes a name or description of the first communication address or a name or description of the first entity.

12. The method of claim 11, further comprising:

determining a relevance measure between the input content and the first communication address or the first entity based on the relation;

when two or more communication addresses or entities are output or displayed, ranking the two or more communication addresses or entities based on the respective relevance measure.

13. The method of claim 11, wherein the first communication address is in the form of a character string, and the second term is a sub-string of the first communication address, wherein the relation is identified by matching the first term with the second term.

14. The method of claim 11, wherein the second term is a term in a name or description associated with the first communication address or the first entity, wherein the relation is identified by matching the first term with the second term.

15. The method of claim 14, further comprising:

identifying an element in a context of the first term, wherein the element indicates that the first term represents an recipient name associated with the first communication address, wherein the element comprises one or more terms or characters including punctuation marks surrounding the first term.

16. The method of claim 11, wherein the second term is a term that is pre-specified by a user to be associated with the first communication address, wherein the relation is identified by matching the first term with the second term.

17. The method of claim 11, wherein the first term is identified as a term representing a topic or topic category in the input content, wherein the second term is identified as a term indicating an interest or activity associated with the topic or topic category, wherein the first term does not match the second term.

18. The method of claim 11, further comprising:

selecting a communication tool and suggesting the communication tool to the user for sending the input content to the first communication address, wherein the communication tool is selected based on an association between the first communication address and the communication tool, wherein the communication tool comprises at least an email composing tool, a chat tool, or a sharing tool;

suggesting or providing the communication tool to the user.