Member communication reply score calculation

Patent metadata

Field	Value
Publication number	US-9569735-B1
Application number	US-201514975756-A
Country	US
Kind code	B1
Filing date	Dec 19, 2015
Priority date	Dec 19, 2015
Publication date	Feb 14, 2017
Grant date	Feb 14, 2017

Abstract

In an example embodiment, a supervised machine learning algorithm is used to train a communication reply score model based on an extracted first set of features and second set of features from social networking service member profiles and activity and usage information. When a plurality of member search results is to be displayed, for the member identified in each of the plurality of member search results, the member profile corresponding to the member is parsed to extract a third set of one or more features from the member profile, activity and usage information pertaining to actions taken by the members on the social networking service is parsed to extract a fourth set of one or more features, and the extracted third set of features and fourth set of features is inputted into the communication reply score model to generate a communication reply score, which is displayed visually to a searcher.

First claim

What is claimed is:

1. A computer-implemented method for providing an indication of a probability that a member of a social networking service will respond to an electronic communication sent via the social networking service, the method comprising:
retrieving a plurality of sample member profiles of members of the social networking service, a plurality of sample member labels, and activity and usage information pertaining to actions taken by those members on the social networking service;
for each sample member profile:
parsing the sample member profile to extract a first set of one or more features from the sample member profile and parsing the activity and usage information pertaining to actions taken by those members on the social networking service to extract a second set of one or more features;
feeding the sample member labels, extracted first set of features and second set of features into a supervised machine learning algorithm to train a communication reply score model based on the extracted first set of features and the second set of features;

obtaining a plurality of member search results produced by actions performed in a user interface, each member search result identifying a member of the social networking service;
for the member identified in each of the plurality of member search results:
parsing a member profile corresponding to the member to extract a third set of one or more features from the member profile and parsing activity and usage information pertaining to actions taken by the members on the social networking service to extract a fourth set of one or more features;
inputting the extracted third set of features and fourth set of features into the communication reply score model to generate a communication reply score reflecting a probability that the member will respond to an email communication from a searcher; and

presenting the member search results visually in the user interface, with each member search result being presented with a visual indication of the corresponding member's communication reply score.

2. The method of claim 1 , wherein the third set of features is identical to the first set of features and the fourth set of features is identical to the second set of features.

3. The method of claim 1 , wherein the obtaining the plurality of member search results includes obtaining an ordering of the plurality of member search results, the ordering based on a ranking of each member search result based on a search algorithm; and
wherein the presenting the member search results causes the member search results to be displayed visually in an order reflecting the ordering, regardless of the communication reply scores of the corresponding members.

4. The method of claim 1 , further comprising:
receiving a selection of one or more members from the user interface as favorites; and
for each of the one or more members selected as favorites, periodically repeating the parsing and inputting for the corresponding member and notifying the searcher if a communication reply score for the corresponding member changes significantly.

5. The method of claim 1 , further comprising:
grouping each member communication reply score into a category based on its relationship to an average communication reply score among a plurality of members; and
wherein the visual indication includes a visual indication of the corresponding grouping for the member communication reply score for the corresponding member.

6. The method of claim 1 , further comprising:
retrieving a plurality of sample searcher member profiles of members of the social networking service, and activity and usage information pertaining to actions taken by those searchers on the social networking service;
for each sample searcher member profile:
parsing the sample searcher member profile to extract a fifth set of one or more features from the sample searcher member profile and parsing the activity and usage information pertaining to actions taken by those searchers on the social networking service to extract a sixth set of one or more features; and
feeding the extracted fifth set of features and sixth set of features into the supervised machine learning algorithm to train the communication reply score model based on the extracted fifth set of features and the sixth set of features.

7. The method of claim 6 , further comprising:
obtaining an identification of the searcher from the user interface;
parsing a member profile corresponding to the searcher to extract a seventh set of one or more features from the member profile and parsing activity and usage information pertaining to actions taken by the searcher on the social networking service to extract an eighth set of one or more features; and
inputting the extracted seventh set of features and eight set of features into the communication reply score model to generate the communication reply score reflecting a probability that the member will respond to an email communication from the searcher.

8. A system comprising:
a non-transitory computer readable medium having instructions stored there on, which, when executed by a processor, cause the system to:
retrieve a plurality of sample member profiles of members of a social networking service, a plurality of sample member labels, and activity and usage information pertaining to actions taken by those members on the social networking service;
for each sample member profile:
parse the sample member profile to extract a first set of one or more features from the sample member profile and parsing the activity and usage information pertaining to actions taken by those members on the social networking service to extract a second set of one or more features;
feed the sample member labels, extracted first set of features and second set of features into a supervised machine learning algorithm to train a communication reply score model based on the extracted first set of features and the second set of features;

obtain a plurality of member search results produced by actions performed in a user interface, each member search result identifying a member of the social networking service;
for the member identified in each of the plurality of member search results:
parse a member profile corresponding to the member to extract a third set of one or more features from the member profile and parsing activity and usage information pertaining to actions taken by the members on the social networking service to extract a fourth set of one or more features;
input the extracted third set of features and fourth set of features into the communication reply score model to generate a communication reply score reflecting a probability that the member will respond to an email communication from a searcher; and

present the member search results visually in the user interface, with each member search result being presented with a visual indication of the corresponding member's communication reply score.

9. The system of claim 8 , wherein the third set of features is identical to the first set of features and the fourth set of features is identical to the second set of features.

10. The system of claim 8 , wherein the obtaining the plurality of member search results includes obtaining an ordering of the plurality of member search results, the ordering based on a ranking of each member search result based on a search algorithm; and
wherein the presenting the member search results causes the member search results to be displayed visually in an order reflecting the ordering, regardless of the communication reply scores of the corresponding members.

11. The system of claim 8 , wherein the computer readable medium further has ins

Assignees

• Linkedin Corp

Inventors

• Zhu Qiang
• Liang Keqing
• Rigano Peter Hume
• Tague Matthew Steven

Classifications

• G06Q10/40

  Business processes related to social networking or social networking services · CPC title

• G06N7/01 Primary

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06N99/005 Primary

  Physics · mapped topic

• G06N7/005

  Physics · mapped topic

• G06F17/3053

  Physics · mapped topic