Browse Prior Art Database

Real-Time Targeted Advertisement System Using Sentiment Analysis

IP.com Disclosure Number: IPCOM000243607D
Publication Date: 2015-Oct-05
Document File: 2 page(s) / 133K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a system to provide real-time targeted advertisements based on sentiment analysis and the location of the user. The system utilizes heart rate data that is collected by a wearable health-monitoring device as an indicator of customer mood, in conjunction with geolocation and the customer profile, to identify and send an appropriate purchasing incentive to the customer.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2

Real-Time Targeted Advertisement System Using Sentiment Analysis

Currently, retailers utilize the proximity of mobile users to certain parts of a physical store, correlated with the customer's stored profile, to send personalized messages about nearby products or displays and subsequently influence purchasing behavior. However, these systems and methods do not consider the emotional state of the user,

which is a significant factor for buying behavior.

For example, if Retailer A detects a spike of interest as Customer B looks at the latest designer boots, followed by a moment of despair after seeing the price, all in real-time, then Retailer A can instantly send a 10% off coupon for the designer boots to Customer
B. This might relieve the negative emotion enough to lead Customer B to purchase the boots.

The novel contribution is a system to provide real-time targeted advertisements based on sentiment analysis and the location of the user. The system utilizes heart rate data,

which is collected by a wearable health-monitoring device, as an indicator of customer mood. This data is then associated with the geolocation of the customer as well as the company's customer profile. Knowing the relative emotional state, location, and customer profile enables the retailer to influence the user's buying behavior in real-time. The retailer can send messages or information designed for the targeted customer. This can include sending special deals directly to the customer's mobile device or providing personalized information on nearby displays (as has been previously done and patented solely based on location/profile). The advantage of the novel approach is that it provides an estimation of mood state, which has been shown to affect consumer

behavior.*

To implement the real-time targeted advertisement system using sentiment analysis in a preferred embodiment:

1. Customer A uses a smart watch that frequently checks heart rate 2. Customer A accepts Retailer X's invitation to opt-into sharing the collected heart rate and geolocation information
3. The associated smart watch sends heart rate data to Customer A's smart phone
4. Customer A's smart phone sends the heart rate data to Retailer X (as is done

when geolocation information is shared)

5. Retailer X uses changes in Customer A's heart rate as a heuristic for Customer

A's mood

6. Retailer X uses a learning algorithm (e.g., Q-learning algorithm) to try different actions based on cus...