Browse Prior Art Database

Method and System for Selecting Products to Display Based on Social Media User-Generated Content

IP.com Disclosure Number: IPCOM000236117D
Publication Date: 2014-Apr-07
Document File: 2 page(s) / 37K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method and system is disclosed for selecting products to display based on social media user-generated content. The method and system analyzes the social media user-generated content within a spatio-temporal boundary and selects content to be displayed on in-store display screen or on a web page.

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

Page 01 of 2

Method and System for Selecting Products to Display Based on Social Media User-

-Generated Content

Generated Content

Disclosed is a method and system for selecting products to display based on social media user-generated content. The method and system enables a user to select content such as, but not limited to, a set of products, advertisements and color schemes to display across one or more screens in either a physical environment or a virtual environment. Here, the physical environment corresponds to, but is not limited to, a retail store and the virtual environment corresponds to, but is not limited to, an online web page. Here, the selection is based on factors such as, but not limited to, sentiments, emotions, moods, topics, intents, products, and colors expressed in a content of messages posted on social media services .

A crawling unit captures the messages across multiple social media services relevant to a clothing domain. Thereafter, data and metadata are loaded and stored for analysis . Here, the messages are either analyzed in real-time or stored and analyzed offline.

A spatio-temporal unit accepts a location and time slice preference from an analyst and groups messages based on the location and time -stamp information. For example, if the analyst wants to see trends by a day and a city, the spatio-temporal unit identifies attributes specified by the analyst in the data, and creates N clusters for N cities per day. A message is placed in a cluster Ni, if the message is originated from a city Ni. The method and system obtains spatial coordinates from the metadata attached to the message and from the content of the message using information extraction techniques for locating spatial entities.

A preference mining unit mines each cluster obtained from the spatio -temporal unit and determines the sentiment of the message for example positive , neutral and negative, topic of the message such as, but not limited to, fashion, weather and pool parties. The preference mining unit also determines specific products , product categories and brands mentioned in the message. Further, the preference mining unit determines the emotions and moods such as, but not limited to, love, anger and happy as expressed in the message. Furthermore, the preference mining unit determines the colors such as , but not limited to, red and blue mentioned in the message, the intents expressed in the message such as, but not limited to, user's desires and user's needs and demographics of an author of the message such as , but not limited to a gender and an age from the message metadata. Thereafter, the preference mining unit selects features that are most representative for the cluster Ni by utilizing all the messages in the cluster Ni .

Preferences are extracted from the social media messages originating from specific locations a...