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

Product Up-Sell, Cross-Sell Based on Social Media Information

IP.com Disclosure Number: IPCOM000246320D
Publication Date: 2016-May-30
Document File: 2 page(s) / 33K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method to combine entity analytics and sentiment on social media to determine the feasibility of a product/service up-sell.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Page 01 of 2

Product Up

Product Up-

Customer up-sell is generally done based on risk (e.g., in credit card industry), past spending habits (e.g., in the retail industry), and other factors. Only recently have sellers begun to use social media information for up-selling. The processes to perform upsell/cross-sell based on social media are in development.

The novel contribution is a method to combine entity analytics and sentiment on social media to determine the feasibility of a product/service up-sell. This method comprises the following processes:


1. Social Media Scraping


2. Entity Detection


3. Sentiment Analysis


4. Temporal Sentiment Aggregation


5. Up-Sell/Cross-Sell Database


6. Sentiment Similarity

1. Social media scraping: An existing customer's social media account(s) is scraped based on information provided at service sign-on. If a user account name does not exist, the system can use contextual computing techniques (i.e., entity analytics) to determine if a certain social media account matches known information entered during service sign -in.

2. Entity Detection: This process locates, extracts and classifies elements in text into pre-defined categories such as product names, customer names, etc.

3. Sentiment Analysis: This process is a form of opinion mining where the result of operation on social media information can yield tristate (i.e., positive, neutral, negative) or numerical (i.e., -1 to 1) results to denote negative/positive sentiment.

4. Temporal Sentiment Ag...