Active user message diet

Patent metadata

Field	Value
Publication number	US-2017372038-A1
Application number	US-201615194300-A
Country	US
Kind code	A1
Filing date	Jun 27, 2016
Priority date	Jun 27, 2016
Publication date	Dec 28, 2017
Grant date	—

Abstract

A system, a machine-readable storage medium storing instructions, and a computer-implemented method are described herein are directed to a Message Diet Engine that generates a pool of messages for a plurality member accounts of a social network service. Each message being of a respective message type from a plurality of message types and targeted to a specific member account. For each respective member account, the Message Diet Engine selects a minimum number of messages, from the pool of messages, targeted to the respective member account that prompts an expected social network activity target and avoids an expected number of complaints. Based on the selected minimum number of messages for each respective member account, the Message Diet Engine identifies a total minimum number of messages, from the pool of messages, to be sent to the plurality of member accounts that prompts an expected total social network activity target and avoids a total expected number of complaints.

First claim

What is claimed is:

1 . A computer system comprising:
one or more hardware processors; and a memory device that stores an instruction set executable by the one or more processors that, when executed, configures the computer system to perform operations comprising: generating a pool of messages for a plurality member accounts of a social network service, each message being of a respective message type from a plurality of message types and targeted to a specific member account; for each respective member account:
selecting a minimum number of messages, from the pool of messages, targeted to the respective member account that prompts an expected social network activity target and avoids an expected number of complaints; and
based on the selected minimum number of messages for each respective member account, identifying a total minimum number of messages, from the pool of messages, to be sent to the plurality of member accounts that prompts an expected total social network activity target and avoids a total expected number of complaints.

2 . The computer system as in claim 1 , further comprising:
collecting historical message data, the historical message data comprising a plurality of previously sent messages in the social network service and a response prompted by each previously sent message.

3 . The computer system as in claim 2 , wherein the response comprises one of an indication of social network activity by a recipient member account in response to message receipt and an indication of a complaint from the recipient member account in response to message receipt.

4 . The computer system as in claim 3 , wherein the complaint comprises one of an unsubscribe action performed by the recipient member account and an identification of the previously sent message as a spam message performed by the recipient member account.

5 . The computer system as in claim 2 , further comprising:
generating, based on the historical message data, an expected social network activity machine learning model and an expected number of complaints machine learning model, wherein the expected social network activity machine learning model returns first output representing a probability that a given set of messages that includes one or more message types will prompt social network activity by a given member account, wherein the expected number of complaints machine learning model returns a number of expected number of complaints from the given member account due to receipt of the given set of messages.

6 . The computer system as in claim 5 , wherein generating, based on the historical message data, an expected social network activity machine learning model and an expected number of complaints machine learning model comprises:
for the expected social network activity machine learning model:
learning a first plurality of regression coefficients, each regression coefficient in the first plurality of regression coefficients representing a degree of importance in which a given message type prompts social network activity by any member account having a given set of a plurality of features, and
for the expected number of complaints machine learning model:
learning a second plurality of regression coefficients, each regression coefficient in the second plurality of regression coefficients representing a degree of importance in which the given message type prompts a complaint by any member account having the given set of a plurality of features.

7 . The computer system as in claim 6 , wherein selecting a minimum number of messages, from the pool of messages, targeted to the respective member account that prompts an expected social network activity target comprises:
identifying, via the expected social network activity machine learning model, from a first plurality of permutations of sets of messages targeted to the respective member account, a first select set of messages that meets or exceeds the expected social network activity target, the first select set of message comprising one or more message types.

8 . The computer system as in claim 7 , wherein selecting a minimum number of messages, from the pool of messages, targeted to the respective member account that avoids an expected number of complaints comprises:
identifying, via the expected number of complaints machine learning model, from a second plurality of permutations of sets of messages targeted to the respective member account, a second select set of messages that avoids an expected number of complaints, the second select set of message comprising one or more message types.

9 . The computer system as in claim 8 , further comprising:
concurrently identifying the first select set of messages with the second select set of messages.

10 . The computer system as in claim 9 , further comprising:
identifying a select minimum number of messages to be sent to the respective member account based upon a convergence of the first select set of messages and the second select set of messages.

11 . A non-transitory computer-readable medium storing executable instructions thereon, which, when executed by a processor, cause the processor to perform operations including:
generating a pool of messages for a plurality member accounts of a social network service, each message being of a respective message type from a plurality of message types and targeted to a specific member account; for each respective member account:
selecting a minimum number of messages, from the pool of messages, targeted to the respective member account that prompts an expected social network activity target and avoids an expected number of complaints; and
based on the selected minimum number of message for each respective member account, identifying a total minimum number of messages, from the pool of messages, to be sent to the plurality of member accounts that prompts an expected total social network activity target and avoids a total expected number of complaints.

12 . The non-transitory computer-readable medium as in claim 11 , further comprising:
collecting historical message data, the historical message data comprising a plurality of previously sent messages in the social network service and a response prompted by each previously sent message.

13 . The non-transitory computer-readable medium as in claim 12 , wherein the response comprises one of an indication of social network activity by a recipient member account in response to message receipt and an indication of a complaint from the recipient member account in response to message receipt.

14 . The non-transitory computer-readable medium as in claim 13 , wherein the complaint comprises one of an unsubscribe action performed by the recipient member account and an identification of the previously sent message as a spam message performed by the recipient member account.

15 . The non-transitory computer-readable medium as in claim 12 , further comprising:
generating, based on the historical message data, an expected social network activity machine learning model and an expected number of complaints machine learning model, wherein the expected social network activity machine learning model returns first output representing a probability that a given set of messages that includes one or more message types will prompt social network activity by a given member account, wherein the expected number of complaints machine learning model returns a number of expected number of complaints from the given member account due to receipt of the given set of messages.

16 . The non-transitory computer-readable medium as in cl

Assignees

• Linkedln Corp

Inventors

• Gupta Rupesh
• Liang Guangeng

Classifications

• G06Q10/40 Primary

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

• G06N7/01

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

• G06N5/01

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title

• G16H50/50

  for simulation or modelling of medical disorders · CPC title

• G16H50/20

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title