Browse Prior Art Database

Tag Recommendation for Social Media Text Content Based on Crowdsourcing

IP.com Disclosure Number: IPCOM000240859D
Publication Date: 2015-Mar-06
Document File: 3 page(s) / 133K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed are a method and system to use crowdsourcing to optimize future tagging by accurately identifying and recommending tags for social media content. The system recommends the most relevant tags for social media content when user is not certain which tags to use.

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

Page 01 of 3

Tag Recommendation for Social Media Text Content Based on Crowdsourcing

With the growing popularity of community driven social media sites such as Facebook* and Twitter*, users share massive amounts of textual content with friends and global users with the world. Social media sites provide tags to categorize content and deliver efficiently deliver it to users. Tags are descriptive keywords that can also be used to query content. For example, an article about California may be tagged with #sunny, #beaches, #Disneyland, #SF.

When publishing content to a social media site, a user might have difficulty determining which tags are most relevant to the content. Currently, social media users have to know or guess the tags that are most relevant to the content, and may only know the popular or trending tags. The drawback is that a user may waste time searching for something appropriate and may end up picking an inaccurate tag.

Currently, there is no known solution for effectively identifying and recommending tags for social media content based on crowdsourcing.

The novel contribution is a method and system to use crowdsourcing to optimize future tagging by accurately identifying tags for social media content. When users tag content with some degree of certainty, the system can then parse the content and correlate keywords to the tags. The system recommends the most relevant tags for social media content

when user is not certain which tags to use. It provides a ranked list of recommended tags based on how other users have tagged similar content. It addition, it can connect words from different languages to enable annotating content in one language with tags from another language. This is a self-learning system increases in accuracy over time.

The following example embodiment illustrates the process for implementing the system for recommending tags for social media content based on crowdsourcing.


1. User A blogs about a new restaurant in which she recently dined while in Paris


2. User A tags the post with #eiffeltower, #restaurant, and #winetasting before publishing it to the web

3. The system parses User A's blog post and associates keywords with the three tags User A used. The system ignores filler text such as 'the', 'a', 'like', 'and', etc. The identified keywords include "France", "modern", "delicious", "cheese", etc. These keywords are represented as color markers in the image below.

4. User B also visits the same restaurant and posts a restaurant review. In it, he tags #eiffeltower, #restaurant, and #ratatouille

5. The system parses through User B's content and associates keywords in the post with the tags. These keywords include "France", "vegetarian", "delicious", "breathtaking views", etc.

1


Page 02 of 3

Figure: Tag database for #eiffeltower


6. Each tag has an associated collection of keywords, which the system aggregated from both User A'...