Browse Prior Art Database

Leveraging sentiment analysis on social network comments to associate a Like

IP.com Disclosure Number: IPCOM000241855D
Publication Date: 2015-Jun-04
Document File: 2 page(s) / 91K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is system that provides an accurate value for a social network response to a post based on both the number of "Like" responses received and the quantity of comments the system determined to be positive.

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

Page 01 of 2

Leveraging sentiment analysis on social network comments to associate a Like

On many social network applications, users have the non-mutually exclusive options for responding posts in the feed/content in the social network (e.g., "Like", comment, or share). A comment could express a positive sentiment; however, the user may have only commented, without clicking "Like" on the post. The Like count therefore does not accurately reflect the amount of positive sentiment toward that post. Since the Like count is the key data point for many decisions in advertising and information gathering, the accuracy of this value is essential. There currently is no known solution to provide that accuracy.

A method is needed to provide an accurate value for a social network's response to a post based on both the number of Like responses received and the quantity of positive comments.

The invention is a system that provides the ability to leverage sentiment analysis of the content as means of categorizing a comment. When the system determines that an evaluated comment is positive in nature, it also verifies whether the user is in the Like count list, meaning the user has also registered a Like for the post. If the system determines that a comment is positive in nature, and it is not already reflected in the Like count, then the system increases the overall Like count. This provides a more accurate reflection of the positive sentiment toward a particular post that may otherwise be lost if the user does not register a Like along with the positive comment.

The novel contributions of the invention include methods to:


• Alter Like counts based on a sentiment analysis of comments

• Register a partial Like based on posted content
• Filter multiple Likes by a single user from multipl...