Techniques for device configuration for commerce messaging using commerce messaging history information

Patent metadata

Field	Value
Publication number	US-10885044-B2
Application number	US-201615350004-A
Country	US
Kind code	B2
Filing date	Nov 11, 2016
Priority date	Apr 11, 2016
Publication date	Jan 5, 2021
Grant date	Jan 5, 2021

Abstract

Various embodiments are generally directed to techniques for device configuration using commerce messaging history information. In one embodiment, an apparatus may comprise a client front-end component operative to receive a client inbox request for a user account from a client device, the user account for a messaging system; and transmit an ordered business promotion contact list to the client device in response to the client inbox request; a business contact list component operative to determine a business promotion contact list for a user account for a messaging system; a predicted interest component operative to determine a predicted business messaging interest for each business contact on the business promotion contact list; and a contact ranking component operative to determine a ranking weight for each business contact on the business promotion contact list based on the predicted business messaging interest for each business contact. Other embodiments are described and claimed.

First claim

The invention claimed is:

1. A computer-implemented method, comprising:
determining a business promotion contact list for an account for a social networking system, the business promotion contact list initially comprising business contacts the account has messaged, and business contacts that friends of the account on the social networking system have messaged, indicating a potential interest in messaging each business contact;
determining a predicted business messaging interest for each business contact based on weights of a linear function applied to a responsiveness rate to messages received by the business contact and an average amount of time for the business contact to respond to received messages, the weights trained based on the responsiveness rate of each business contact, the average amount of time for each business contact to respond to received messages, and a friend engagement measure for each business contact, wherein the friend engagement measure is based at least in part on whether the friends of the account have liked a page for the business contact on the social networking system;
determining a ranking weight for each business contact based on the predicted business messaging interest for each business contact; and
ordering the business promotion contact list for display for the account based on the determined ranking weights.

2. The method of claim 1 , wherein the predicted business messaging interest is further based on the friend engagement measure for each business contact.

3. The method of claim 1 , wherein a linear regression modifies the weights of the linear function.

4. The method of claim 1 , further comprising:
receiving a plurality of business features for each business contact on the business promotion contact list;
receiving a plurality of profile features for a plurality of accounts of the social networking system;
receiving a messaging history for the social networking system;
generating a predicted business messaging interest model for the social networking system based on a machine learning analysis of the messaging history, the predicted business messaging interest model based on the plurality of profile features and the plurality of business features; and
determining the predicted business messaging interest for each business contact on the business promotion contact list based on the predicted business messaging interest model.

5. The method of claim 4 , wherein the predicted business messaging interest model predicts one or more of user interest in beginning a messaging interaction with a business, user interest in requesting a service via the messaging interaction with the business, and user satisfaction from requesting the service via the messaging interaction with the business.

6. The method of claim 4 , wherein the predicted business messaging interest model predicts one or more of user interest in subscribing with a business, user time spent reading subscription messages from the business, user responsiveness to subscription notifications for the business, and user blocking of the business.

7. The method of claim 1 , the account associated with one or more friends with the social networking system, further comprising:
generating a seed business promotion contact list for the account, wherein the seed business promotion contact list is based on a list of business contacts the account has messaged with via the social networking system and a list of business contacts the friends of the account have messaged with via the social networking system; and
determining feature weights for the seed business promotion contact list based on a machine learning analysis of a messaging history for the social networking system; and
expanding the seed business promotion contact list to generate the business promotion contact list by including additional business contacts with similar features to the business contacts on the seed business promotion contact list.

8. An apparatus, comprising:
a processor circuit on a device;
memory, for execution on the processor circuit, the memory containing instructions implementing:
a client front-end component operative to:
receive a client inbox request for an account from a client device, the account for a social networking system; and
transmit an ordered business promotion contact list to the client device in response to the client inbox request, the ordered business promotion contact list ordered for display for the account based on a determined ranking weight for each business contact on the ordered promotion contact list;

a business contact list component operative to:
determine a business promotion contact list for the account, the business promotion contact list initially comprising business contacts the account has messaged, and business contacts that friends of the account on the social networking system have messaged, indicating a potential user interest in messaging each business contact;

a predicted interest component operative to:
determine a predicted business messaging interest for each business contact based on weights of a linear function applied to a responsiveness rate to messages received by the business contact and an average amount of time for the business contact to respond to received messages, the weights trained based on the responsiveness rate of each business contact, the average amount of time for each business contact to respond to received messages, and a friend engagement measure for each business contact, wherein the friend engagement measure is based at least in part on whether the friends of the account have liked a page for the business contact on the social networking system; and

a contact ranking component operative to:
determine the ranking weight for each business contact on the business promotion contact list based on the predicted business messaging interest for each business contact.

9. The apparatus of claim 8 , wherein the predicted business messaging interest is further based on the friend engagement measure for each business contact.

10. The apparatus of claim 8 , wherein a linear regression modifies the weights of the linear function.

11. The apparatus of claim 8 , further comprising:
the predicted interest component operative to receive a plurality of business features for each business contact on the business promotion contact list; receive a plurality of profile features for a plurality of accounts of the social networking system; receive a messaging history for the social networking system; generate a predicted business messaging interest model for the social networking system based on a machine learning analysis of the messaging history, the predicted business messaging interest model based on the plurality of profile features and the plurality of business features; and determine the predicted business messaging interest for each business contact on the business promotion contact list based on the predicted business messaging interest model.

12. The apparatus of claim 11 , wherein the predicted business messaging interest model predicts one or more of user interest in beginning a messaging interaction with a business, user interest in requesting a service via the messaging interaction with the business, and user satisfaction from requesting the service via the messaging interaction with the business.

13. The apparatus of claim 11 , wherein the predicted business messaging interest model predicts one or more of user interest in subscribing with a business, user time spent reading subscription messages from the business, user responsiveness to subscription notifications for the business, and user blocking of the business.

Assignees

• Facebook Inc

Inventors

• Zhang Edward Young
• Ding Weiwei
• Wang Qian
• Chen Tsung-Chuan
• Liu Guangyuan

Classifications

• G06F16/24578 Primary

  using ranking · CPC title

• H04W4/12 Primary

  Messaging; Mailboxes; Announcements · CPC title

• H04L67/53

  using third party service providers · CPC title

• H04L65/611

  for multicast or broadcast (systems for broadcast or conference H04L12/18; arrangements for broadcast or distribution combined with broadcast H04H20/00; arrangements for broadcast applications with a direct linkage to broadcast information or to broadcast space-time H04H60/00; selective distribution of broadcast services, e.g. multimedia broadcast multicast service [MBMS], H04W4/06) · CPC title

• H04L51/52

  for supporting social networking services · CPC title