TransparencyEngine – Finding Influencers and Their Tribes


TransparencyEngine (previously called "GalaxyScope") is a new powerful and easy-to-use tool to find the most influential people about any topic on social media. 

You can try out TransparencyEngine here. (You will need a Twitter account and have to login with it.)






In an earlier post I described how to use Condor to run queries for TransparencyEngine. While Condor is powerful in finding hidden trends and influencers on social media such as Twitter, Blogs, Wikipedia, and Web sites and online forums, it has quite a steep learning curve. If one is not a data scientist or computer savvy otherwise, Condor can be quite challenging to extract the results one is looking for.

We therefore have developed a Web version of TransparencyEngine described in this post as a much simpler to use tool.  TransparencyEngine has been created by my colleague Joao Marcos de Oliveira as an easier-to-use alternative to Condor. Other than Condor it is embedded into the Web browser, which means that by just entering some key words in a browser window, and clicking a search button one will be able to get a result.

TransparencyEngine consists of three components, a People Map, a Tribe Map, and a Topic Map. The people map will create a social network for any person who is either in Wikipedia or in Twitter.  In the example below I am calculating the social network for Roger Federer, entering his name into the search box.




I then select the Wikipedia pages and Twitter accounts, which are about Roger Federer, and click on the “Get Network” button., resulting in the social network of Roger Federer, leading to a network which reads like a “who is who” among the top tennis players of the world.




The Topic map will create a tribe of people about any topic. It uses both Twitter and Wikipedia as its main sources for finding the influencers about the topics, which means that it will find concepts that have either been spoken about on Twitter, or that have one or more Wikipedia pages. Similarly, influencers will be found from among Twitter users or if they have their own people page on Wikipedia. In other words, hidden influencers who do not tweet and do not have a page on Wikipedia will remain hidden, as the TransparencyEngine Topic mapping system will not be able to find them. The screen shot below shows the start screen of the topic map



I will illustrate creating a topic map by finding trends and trendsetters about OLED  and LG, which is a major manufacturer of OLED displays.

First I define a new tribe called “OLED” in tribefinder and add the Wikipedia pages about “OLED” “LG”, “Samsung” and “Samsung Electronics” as input for the search for tribe members.


This will add a new topic called “OLED” to the Topic maps. Clicking on the tab “Members” will show the tribe members of the “OLED” tribe. Initially it will be empty. Clicking on the tab “”Profile Description” in the search menu at left allows me to search for Twitter users who have the keywords “OLED” and “LG” in their Twitter profile description.




Similarly, Twitter users can be searched by keywords they are tweeting about. These users are then shown in the main part of the window, and can selectively be added to the tribe. In addition, Twitter users can also be searched among their followers and friends of the people already in the tribe (friends are bi-directional links in Twitter, where people are following each other).

Once the tribe has been created, it can be shown as a network. For this network, the tribe members are connected by their most popular Twitter friends.
The picture below shows the tribe created by selecting the most important Twitter profiles from the search in tweets and in Twitter profile descriptions for the keywords “OLED” and “LG”.


At the same time we can also look at the tag cloud by clicking on the tab “Hashtags” which is extracted from the tweets collected about the search terms, in this case the words “OLED” and “LG”.



In the influencer network shown above, the Twitter profile theMrMobile is one of the most central ones. It is the Twitter account of Michael Fisher, who, according to his Twitter profile description,  produces “Mobile technology videos for the discerning gadget enthusiast”.

Entering the Twitter handle “theMrMobile” into the Tribe finder menu below shows the positioning of the tweets of theMrMobile into 16 predefined tribes belonging to the four categories “alternative realities”, “ideologies”, “lifestyle” and “recreation”. theMrMobile is categorized as nerdish, capitalist, yolo lifestyle, and sporty.



TransparencyEngine's topic view and tribe finder quickly identify the key topics around OLED, such as the smartphone displays of Apple and Samsung,  Clicking on a person in the network view, or on a tag in the hashtag view, will bring up a separate window to show the term in the original context.

The link above is to the alpha version, which  most likely still has quite a few bugs. Please test it out, your feedback is most welcome!

Comments

Popular posts from this blog

Coolhunts for 4/16/07 to 5/11/07

Scientists, Monks and Bankers - It's All About Love

If good people make AI, can AI make good people?