Browse Prior Art Database

A method of suggesting hashtags for a tweet based on it's time, content and author.

IP.com Disclosure Number: IPCOM000234626D
Publication Date: 2014-Jan-23
Document File: 3 page(s) / 96K

Publishing Venue

The IP.com Prior Art Database

Abstract

In a social media tool environment, a method is described for suggesting appropriate contextual metadata (or hashtags). The solution offers users intelligent hashtag selection guidance when constructing messages. A curated list of the most relevant and widely used hashtags culled in real time from global and social usage patterns is presented to the user.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 45% of the total text.

Page 01 of 3

A method of suggesting hashtags for a tweet based on it's time, content and author.

Social media tools (e.g. twitter, facebook etc.) group conversations around contextual metadata, known as hashtags, but very little is done to encourage and assist users in applying hashtags to their own messages so that they can contribute to the wider conversation around their subject. In most cases, a user has to construct their own "best guess" hashtag or pick from the few that are currently trending. In the former case, a "best guess" will often result in many users creating

many varied hashtags; this precludes them from being grouped together as one conversation, lessening their usefulness to the user, the rest of the tool's ecosystem and any companies performing downstream analysis. Choosing from a trending list can be even less useful, as the number of hashtags presented is extremely limited and additionally restricted to super subjects (e.g. major global and national events)

which may not be relevant to the content of the user's message. As well as the manual solutions mentioned previously, there are several existing mechanisms

which operate in a similar area. These are listed below with their description and associated drawbacks:


• hashtagify.me - allows you to search for a hashtag and find the top 10 most related ones. This solution relies heavily on the user to select an initial starting hashtag.


• Twitter - provides a hashtag suggestion system but it is not used when composing a tweet.


• hashtags.org - is similar to hashtagify.me in that you are required to select a starting hashtag.

• Twubs - is a directory of hashtags, rather a recommendation system.

The solution presented herein offers users intelligent hashtag selection

guidance when constructing messages to encourage their use and the users' participation in the wider discussion. It will present the user with a curated list of the most relevant and widely used hashtags culled in realtime from global and social usage patterns. Businesses employing data mining techniques on social media gain extra value from analysing this broader, more relevant data pool.

The solution presented here will make hashtag suggestions to the author as they are composing their tweet. These suggestions will be relevant to the author in three

ways:


1. The solution will search their social circles and give some weight to results found there, which means the suggestions are relevant in that they come from the people who are important to the author.


2. The solution will search recent activity on Twitter, which means that the

suggestions are relevant in that they come from what is happening right now in the

world.


3. The solution will search based on the content of the tweet, which means that the

suggestions are relevant to what the author is wanting to talk about.

    The novelty of this solution lies within these three items, in that the algorithm is designed in such a way that the suggestions provided a...