Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

A Method to Dynamically Provide Relevant Content Derived from Heterogeneous Data Source-Based User Preferences and Activity Patterns

IP.com Disclosure Number: IPCOM000244162D
Publication Date: 2015-Nov-17
Document File: 3 page(s) / 55K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a system to learn and aggregate user preferences and activity patterns through heterogeneous data sources in order to deliver relevant content to a target recipient. The approach combines (at least) three heterogeneous data sources as input: sensor-based smart Internet of Things (IoT) system, the user’s Global Positioning System (GPS) location traces, and the user’s electronic calendar entries.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 3

A Method to Dynamically Provide Relevant Content Derived from Heterogeneous Data Source -

-Based User Preferences and Activity Patterns

Based User Preferences and Activity Patterns

Organizations often send specific content (e.g., advertisements, coupons, recommendations, etc.) to electronic communications users. The delivered content is often not useful or relevant to the recipient . A system or method is needed to better consider user preferences, conditions, activity patterns, etc. in order to identify the appropriate recipients for specific content.

The novel solution is a system to learn and aggregate user preferences and activity patterns through heterogeneous data sources in order to deliver relevant content to a target recipient. The core novelty is the function of using the combination of (at least) the following three heterogeneous data sources as input :

• Sensor-based smart Internet of Things (IoT) system • User's Global Positioning System (GPS) location traces
• User's electronic calendar entries

The method to dynamically provide relevant content to a given user consists of the following steps:

1. Continually obtain data from three or more heterogeneous data sources , these sources include, but are not limited to:

A. Sensors and cameras using in a smart IoT system

    B. User's GPS location traces
C. User's electronic calendar
2. Learn user activity patterns from the heterogeneous data
3. Derive user preferences from the heterogeneous data
4. Aggregate the user preferences and activity patterns

5. Apply data to deliver relevant content to the user

6. Repeating steps 1-5 to update the data, patterns, preferences, and deliver the currently relevant content

Figure: A high-level process flow of the proposed approach

1


Page 02 of 3

Use Case #1


1. Learning Activity Patterns and Preferences:

A. By analyzing the GPS location traces of the user, the process (i.e. system users) can infer valuable patterns of user's daily life. For example,

• Every Saturday evening, the user goes to a pub

• Every Wednesday, the user goes to a pizza restaurant

B. The IoT system consisting of sensing devices in the user's...