User/group servicing based on deep network analysis

Patent metadata

Field	Value
Publication number	US-2019019221-A1
Application number	US-201715649272-A
Country	US
Kind code	A1
Filing date	Jul 13, 2017
Priority date	Jul 13, 2017
Publication date	Jan 17, 2019
Grant date	—

Abstract

Content is selectively provided to users of mobile devices within a venue including an on-site wireless network. User authorization requests and/or user account registration data are transmitted to the on-site wireless network from mobile devices within the venue. Attributes such as user interests and professions, which comprise inferred user profiles, are obtained using the network traffic data. Identities of mobile devices are established based on a combination including two or more of network identifiers, mobile device signatures, and browser signatures. The inferred user profiles are correlated with the mobile device identities. The inferred user profiles are aggregated into user profile groups and then matched with a content provider's intended target profiles. Content is transmitted to the mobile devices corresponding to the intended target profiles and based on correlation of the inferred user profiles with identities of the devices. Inferred user profiles may be verified using social and/or geographical data.

First claim

1 .- 11 . (canceled)

12 . A system for providing content regarding users of mobile devices, comprising:
an on-site wireless network configured to process user authorization requests and/or user account registration data from mobile devices within a venue and facilitate communication between authorized mobile devices within the venue and content providers; a network analytics engine configured to:
receive hashed user identification data and network traffic data from the on-site wireless network,
establish identities of the mobile devices based on a combination comprising at least two of:
network identifiers, mobile device signatures, and browser signatures,
obtain inferred user profiles from uniform resource locators (URLs) and Internet Protocol (IP) headers using the network traffic data,
correlate the inferred user profiles with the identities of the mobile devices,
aggregate the inferred user profiles into user profile groups, and match the user profile groups with a content provider's intended target profile,

the on-site wireless network being further configured for transmitting content to selected ones of the mobile devices based on one or more of the user profile groups matching the content provider's intended target profile and based on the correlation of the inferred user profiles with the identities of the mobile devices.

13 . The system of claim 12 , wherein the on-site wireless network comprises a Wi-Fi network.

14 . The system of claim 13 , wherein the network analytics engine is further configured to:
extract social data from one or more social media sites, obtain social profiles from the social data, determine which of the inferred user profiles obtained from the traffic data are consistent with the social profiles to obtain verified inferred user profiles, and aggregate the verified inferred user profiles into the user profile groups.

15 . The system of claim 12 , wherein the network analytics engine further includes a data content analyzer configured to determine which of the URLs are associated with web pages viewable to users of the mobile devices and provide a first output based on the viewable pages.

16 . The system of claim 15 , wherein the data content analyzer is further configured to conduct domain pulls and IP reverse lookups using the IP headers to identify one or more of destination regions, service types and apps.

17 . The system of claim 16 , wherein the network analytics engine further includes an insight correlation engine configured to process the first output from the data content analyzer, the insight correlation being configured to:
perform web page content analysis by identifying one or more ranked keywords, and correlate identified ranked keywords to infer mobile device user interests.

18 . The system of claim 17 , wherein the insight correlation engine is further configured to obtain content type reflecting user interests based on the first output from the data content analyzer by:
performing a keyword search on a URL text string, and conducting URL text parsing.

19 . The system of claim 18 , wherein the networks analytics engine further includes a profile and trend inferencer, the insight correlation engine being configured to provide output data to the profile and trend inferencer, the profile and trend inferencer being configured to match the output data to a set of user profile models to obtain the inferred user profiles.

20 . A non-transitory computer readable medium comprising computer executable instructions which when executed by a computer cause the computer to perform a method of providing content regarding users of mobile devices within a venue comprising an onsite wireless network, the method comprising:
processing user authorization requests and/or user account registration data transmitted to the on-site wireless network from mobile devices within the venue; obtaining hashed user identification data and network traffic data from the on-site wireless network, the network traffic data including uniform resource locators (URLs) and Internet Protocol (IP) headers; establishing identities of the mobile devices based on a combination comprising at least two of: network identifiers, mobile device signatures, and browser signatures; obtaining inferred user profiles based on the URLs and IP headers obtained from the network traffic data; correlating the inferred user profiles with the identities of the mobile devices; aggregating the inferred user profiles into user profile groups that match one or more user profile models; matching the user profile groups with a content provider's intended target profile; and
transmitting content to selected ones of the mobile devices based on one or more of the user profile groups matching the content provider's intended target profile and based on the correlation of the inferred user profiles with the identities of the mobile devices.

Assignees

• IBM

Inventors

• Agrawal Bhavna
• Anerousis Nikolaos
• Schaffa Frank A
• Xiao Jin

Classifications

• G06Q10/40

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

• G06Q30/0269 Primary

  based on user profile or attribute · CPC title

• G06N20/00

  Machine learning · CPC title

• G06N5/04

  Inference or reasoning models · CPC title

• G06Q50/01

  Physics · mapped topic