Enhancement of machine learning techniques for an electronic message system

• US 9,313,166 B1
• Filed: 04/29/2015
• Issued: 04/12/2016
• Est. Priority Date: 04/29/2015
• Status: Active Grant

• Current Assignee: XANT Inc.
• Original Assignee: Insidesales.com
• Inventors: Zeng, Xinchuan
• Primary Examiner(s): Chen, Alan
• Application Number:
  US14/699,990
• Time in Patent Office: 349 Days
• US Class Current: 1/1
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: H04L 51/02 using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
• CPC Class Codes: H04L 51/214 using selective forwarding

Abstract

Techniques are described herein for classifying an electronic message with a particular project from among a plurality of projects. In some embodiments, first and second users associated with the electronic message are identified, and one or more first projects associated with the first user and one and more second projects associated with the second user are determined. Projects that are in common between the first projects and the second projects are determined. When only a single project is in common, the electronic message is associated with the single project. When more than a single project is in common, features associated with each of the projects found to be in common are analyzed by a machine learning model to determine the most likely project to associate with the electronic message from among the projects found to be in common.

Claims

1. A system comprising:

a data store that stores data that indicates which users of a plurality of users are associated with which projects of a plurality of projects;

one or more computing devices, wherein one or more first computing devices of the one or more computing devices are operatively coupled to the data store and programmed to;

identify a first user and a second user associated with a first electronic message;

based on the data from the data store, determine one or more first projects that are associated with the first user;

based on the data from the data store, determine one or more second projects that are associated with the second user;

make a determination as to whether only a single project is common between the one or more first projects and the one or more second projects;

respond to a determination that only a single project is common between the one or more first projects and the one or more second projects by storing, in the data store, an association between the first electronic message and the single project.

2. The system of claim 1, wherein:

only a single project is common between the one or more first projects and the one or more second projects;

the first user is associated with a second electronic message that is not associated with the single project; and

the one or more first computing devices are further programmed to respond to a request from a client computer for one or more electronic messages associated with the single project by sending to the client computer only those electronic messages that are associated with the single project, wherein the electronic messages associated with the single project include the first electronic message but not the second electronic message.

3. The system of claim 1, wherein:

the data store or the one or more computing devices includes a machine learning model;

the one or more first computing devices of the one or more computing devices is programmed to respond to a determination that more or less than a single project is common between the one or more first projects and the one or more second projects by;

collecting one or more first features associated with each of the more or less than the single project in common between the one or more first projects and the one or more second projects;

for each project in the more or less than the single project in common between the one or more first projects and the one or more second projects, generating a project score based on the machine learning model and the one or more first features; and

determining that a particular project has a higher project score than any other project in the more or less than the single project in common between the one or more first projects and the one or more second projects, and in response, associating the first electronic message with the particular project.

4. The system of claim 1, wherein:

the data store or the one or more computing devices includes a machine learning model;

the one or more first computing devices of the one or more computing devices is programmed to respond to a determination that more or less than a single project is common between the one or more first projects and the one or more second projects by;

collecting one or more first features associated with the first electronic message;

for each project in the plurality of projects, generating a project score based on the machine learning model and the one or more first features; and

determining that a particular project has a higher project score than any other project in the plurality of projects, and in response, associating the first electronic message with the particular project.

5. The system of claim 4, wherein:

the data store includes a plurality of electronic messages;

one or more second computing devices of the one or more computing devices are programmed to generate the machine learning model by, for each electronic message of the plurality of electronic messages stored in the data store;

collecting one or more electronic messages features associated with the electronic message; and

training the machine learning model based on the one or more electronic messages.

6. The system of claim 5, wherein at least a first computing device in the one or more first computing devices is also one of the one or more second computing devices.

7. The system of claim 4, wherein:

the data store includes a plurality of electronic messages, wherein each electronic message in the plurality of electronic messages is associated with a project of the plurality of projects;

one or more second computing devices of the one or more computing devices are programmed to generate the machine learning model by, for each electronic message of the plurality of electronic messages;

collecting one or more project features associated with the project that is associated with the electronic message; and

training the machine learning model based on the one or more project features.

8. The system of claim 4, wherein:

the data store includes a plurality of electronic messages, wherein each message in the plurality of electronic messages is associated with a project of the plurality of projects;

one or more second computing devices of the one or more computing devices are programmed to generate the machine learning model by, for each electronic message of the plurality of electronic messages;

collecting one or more matching features, wherein each feature in the one or more matching features is associated with the electronic message and the project associated with the electronic message; and

training the machine learning model based on the one or more matching features.

9. The system of claim 4, wherein:

the data store includes a plurality of electronic messages, wherein each message in the plurality of electronic messages is associated with a project of the plurality of project;

one or more second computing devices of the one or more computing devices are programmed to generate the machine learning model by, for each electronic message of the plurality of electronic messages;

collecting one or more similar features, wherein each feature in the one or more similar features is associated with the electronic message and is similar to a feature associated with the project associated with the electronic message; and

training the machine learning model based on the one or more similar features.

10. The system of claim 1, wherein:

the data store or the one or more computing devices includes a machine learning model;

the one or more first computing devices of the one or more computing devices is programmed to respond to a determination that more or less than a single project is common between the one or more first projects and the one or more second projects by;

collecting one or more first features associated with the first electronic message;

for each project in the plurality of projects, generating a project score based on the machine learning model and the one or more first features;

determining an particular order of the plurality of projects based on the project score for each project in the plurality of projects;

causing a client computer to present the plurality of projects in the particular order to a user;

the one or more first computing devices of the one or more computing devices is programmed to receive input from the user indicating that a particular project should be associated with the first electronic message, and in response, associate the first electronic message with the particular project.

11. A method comprising:

identifying a first user and a second user associated with a first electronic message;

based on data from a data store that indicates which users of a plurality of users are associated with which projects of a plurality of projects, determining one or more first projects that are associated with the first user;

based on the data from the data store, determining one or more second projects that are associated with the second user;

determining whether only a single project is in common between the one or more first projects and the one or more second projects;

in response to the determination that only a single project is in common between the one or more first projects and the one or more second projects, storing, in the data store, an association between the first electronic message and the single project.

12. The method of claim 11, further comprising:

wherein only a single project is in common between the one or more first projects and the one or more second projects and the first user is associated with a second electronic message that is not associated with the single project,

responding to a request from a client computer for one or more electronic messages associated with the single project by sending to the client computer only those electronic messages that are associated with the single project, wherein the electronic messages associated with the single project include the first electronic message but not the second electronic message.

13. The method of claim 11, further comprising:

responding to a determination that more or less than a single project is in common between the one or more first projects and the one or more second projects by;

collecting one or more first features associated with each of the more or less than the single project in common between the one or more first projects and the one or more second projects;

for each project in the more or less than the single project in common between the one or more first projects and the one or more second projects, generating a project score based on a machine learning model and the one or more first features; and

determining that a particular project has a higher project score than any other project in the more or less than the single project in common between the one or more first projects and the one or more second projects, and in response, associating the first electronic message with the particular project.

14. The method of claim 11, further comprising:

responding to a determination that more or less than a single project is common between the one or more first projects and the one or more second projects by;

collecting one or more first features associated with the first electronic message;

for each project in the plurality of projects, generating a project score based on a machine learning model and the one or more first features; and

determining that a particular project has a higher project score than any other project in the plurality of projects, and in response, associating the first electronic message with the particular project.

15. The method of claim 14, further comprising:

generating the machine learning model by, for each electronic message of a plurality of electronic messages stored in the data store;

collecting one or more electronic messages features associated with the electronic message; and

training the machine learning model based on the one or more electronic messages.

16. The method of claim 14, further comprising:

generating the machine learning model by, for each electronic message of a plurality of electronic messages stored in the data store, wherein each electronic message in the plurality of electronic messages is associated with a project of the plurality of projects;

collecting one or more project features associated with the project that is associated with the electronic message; and

training the machine learning model based on the one or more project features.

17. The method of claim 14, further comprising:

generating the machine learning model by, for each electronic message of a plurality of electronic messages stored in the data store, wherein each message in the plurality of electronic messages is associated with a project of the plurality of projects;

collecting one or more matching features, wherein each feature in the one or more matching features is associated with the electronic message and the project associated with the electronic message; and

training the machine learning model based on the one or more similar features.

18. The method of claim 14, further comprising:

generating the machine learning model by, for each electronic message of a plurality of electronic messages stored in the data store, wherein each message in the plurality of electronic messages is associated with a project of the plurality of project;

collecting one or more similar features, wherein each feature in the one or more similar features is associated with the electronic message and is similar to a feature associated with the project associated with the electronic message; and

training the machine learning model based on the one or more similar features.

19. The method of claim 11, further comprising:

responding to a determination that more or less than a single project is in common between the one or more first projects and the one or more second projects by;

collecting one or more first features associated with the first electronic message;

for each project in the plurality of projects, generating a project score based on a machine learning model and the one or more first features;

determining an particular order of the plurality of projects based on the project score for each project in the plurality of projects;

causing a client computer to present the plurality of projects in the particular order to a user;

receiving input from the user indicating that a particular project should be associated with the first electronic message;

in response to the input, associating the first electronic message with the particular project.