Identifying and correlating electronic mail messages

• US 8,423,616 B2
• Filed: 05/03/2007
• Issued: 04/16/2013
• Est. Priority Date: 05/03/2007
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Pouzin, Dominic J., Smith, Ilana Rae
• Primary Examiner(s): Patel, Haresh N
• Assistant Examiner(s): LE, CHAU D
• Application Number:
  US11/799,990
• Publication Number:
  US 20080275957A1
• Time in Patent Office: 2,175 Days
• Field of Search: 709/206, 709/207, 709/238
• US Class Current: 709/206
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: H04L 51/216 Handling conversation history, e.g. grouping of messages in sessions or threads

Abstract

Technologies are described herein for identifying and correlating e-mail messages. The contents of the subject and recipients fields of e-mail messages received at and sent from a computing system are hashed and stored in a database. When an incoming e-mail message is received at the computing system, the contents of its subject and recipients fields are also hashed. A search is then made of the database to locate e-mail messages identified in the database that have a sufficient number of subject and address hash values that match the subject and address hash values generated for the incoming e-mail message so that the messages may be correlated. A correlation is made between the incoming e-mail message and the best matching e-mail message located during the search of the database.

Claims

1. A computer-implemented method for identifying and correlating an incoming electronic mail (e-mail) message to a computing system with an outbound e-mail message previously transmitted by the computing system and referenced in a database stored by the computing system, the method comprising performing computer-implemented operations for:

processing outbound e-mail messages from the computing system by identifying data stored in a subject field and an address field of the outbound e-mail messages and storing secure hash values generated from a secret key in the database for the data stored in the subject field and the address field of the outbound e-mail messages; and

processing an incoming e-mail message to the computing system by determining whether hash values of data stored in a subject field excluding one or more leading words and an address field of the incoming e-mail message sufficiently match the secure hash values stored in the database for the corresponding subject field and address field of the outbound e-mail messages to correlate the incoming e-mail message with one or more of the outbound e-mail messages, and correlating the incoming e-mail message with one or more of the outbound e-mail messages in response to determining that the hash values of data stored in the subject field excluding one or more leading words and the address field of the incoming e-mail message sufficiently match the secure hash values stored in the database for the corresponding subject field and address field of one or more previously transmitted outbound e-mail messages.

2. The method of claim 1, wherein the one or more leading words comprise one or more words of a certain format and wherein the address field of the incoming e-mail message comprises a recipient field.

3. A computer-implemented method for identifying and correlating an incoming electronic mail (e-mail) message to a computing system with one of a plurality of e-mail messages sent from the computing system, each of the e-mail messages having content stored in a subject field and identifying one or more recipients by an e-mail address, the method comprising performing computer-implemented operations for:

processing e-mail messages sent from the computing system by tokenizing the content in the subject field of the plurality of e-mail messages sent from the computing system into one or more tokens, generating a subject hash value for each of the tokens, generating an address hash value for each of the e-mail addresses of the recipients of the plurality of e-mail messages sent from the computing system, and storing the subject hash values and the address hash values in a database; and

processing the incoming e-mail message by tokenizing a subset of the content in a subject field of the incoming e-mail into one or more tokens message, wherein the subset excludes one or more leading words in the subject field, generating a subject hash value for each of the tokens for the subset of the content in the subject field of the incoming e-mail message, generating an address hash value for each of the e-mail addresses of recipients of the incoming e-mail message, and correlating the incoming e-mail message to one or more of the plurality of e-mail messages sent from the computing system using the subject hash values and address hash values generated for the incoming e-mail message and the subject hash values and address hash values stored in the database.

4. The method of claim 3, wherein storing the subject hash values and the address hash values in the database comprises storing a record in the database for each of the plurality of e-mail messages sent from the computing system, each record comprising an identifier for an e-mail message, the subject hash values for an e-mail message, and the address hash values for an e-mail message.

5. The method of claim 4, wherein correlating the incoming e-mail message to one of the plurality of e-mail messages sent from the computing system using the subject hash values and address hash values generated for the incoming e-mail message and the subject hash values and address hash values stored in the database comprises:

performing a search of the database for records stored therein having at least a first threshold number of address hash values that match the address hash values generated for the incoming e-mail message;

identifying zero or more candidate e-mail messages by searching the results of the database search for records having at least a second threshold number of subject hash values that match the subject hash values generated for the incoming e-mail message; and

correlating the incoming e-mail message to one of the candidate e-mail messages.

6. The method of claim 5, wherein correlating the incoming e-mail message to one of the candidate e-mail messages comprises correlating the incoming e-mail message to a most recently modified one of the candidate e-mail messages.

7. The method of claim 6, further comprising providing an indication that the incoming e-mail message is unrelated to any of the plurality of e-mail messages in response to determining that there are zero candidate e-mail messages.

8. The method of claim 7, wherein the database comprises a structured query language (SQL) database.

9. A computer-readable medium comprising a tangible article of manufacture that is not a signal, the computer-readable medium having computer-executable instructions stored thereon which, when executed by a computer, cause the computer to:

process outbound e-mail messages from a computer system by generating subject hash values and address hash values for the outbound e-mail messages and storing a record in a database for each of a plurality of outbound e-mail messages, each record comprising an identifier for the outbound e-mail message, the subject hash values and the address hash values for the outbound e-mail messages; and

to process incoming e-mail messages to the computer system by receiving an incoming e-mail message having a subject field and one or more address fields, each address field having one or more recipient e-mail addresses stored therein, and in response to receiving the incoming e-mail message, to tokenize the subject field of the incoming e-mail message into a plurality of tokens, to generate a subject hash value for each of the tokens, and to generate an address hash value for each of the recipient e-mail addresses, to identify the incoming e-mail message as being related to one or more previously transmitted outbound e-mail messages using the subject hash values and the address hash values for the incoming e-mail message and the subject hash values and the address hash values stored in the database and, in response to identifying the e-mail message as being related to one or more previously transmitted outbound e-mail messages, to correlate the incoming e-mail message to the related one or more previously transmitted outbound e-mail messages by

performing a search of the database for records stored therein having at least a first threshold number of address hash values that match the address hash values generated for the incoming e-mail message,

identifying zero or more candidate e-mail messages by searching the results of the database search for records having at least a second threshold number of subject hash values that match the subject hash values generated for the incoming e-mail message, and

correlating the incoming e-mail message to one of the candidate e-mail messages using the subject hash values and address hash values generated for the incoming e-mail message and the subject hash values and address hash values stored in the database.

10. The computer-readable medium of claim 9 having further computer-executable instructions stored thereon which cause the computer to provide an indication that the incoming e-mail message is unrelated to any of the previously transmitted outbound e-mail messages in response to determining that there are zero candidate e-mail messages.

11. The computer-readable medium of claim 10, wherein correlating the incoming e-mail message to one of the candidate e-mail messages comprises correlating the incoming e-mail message to a most recently modified one of the candidate e-mail messages.

12. The computer-readable medium of claim 11, wherein the database comprises a structured query language (SQL) database.