A Lithium brand
The Klout Score Methodology Secrets Revealed
Francisca
1
Kudos
November 05, 2015

Klout-Data-blog-image.jpg

 

Last week, Lithium Klout data scientists presented an academic paper titled “Klout Score: Measuring Influence Across Multiple Social Networks,” during the IEEE BigData 2015 Conference. The paper gives a look behind the Klout Score and the big data that powers it.

 

The paper is significant because – for the first time ever – it explains the methodology behind the Klout Score.  It details how the Klout data science team employs a scalable system that assigns Klout Scores by processing 45 billion interactions daily, and captures 3,600 different actions that then factor into the Klout Score.

 

Using this system, Klout has scored more than 750 million profiles, giving Lithium one of the largest digital footprints in the world.

 

So, it’s no surprise that Forbes calls the Klout Score “more than just a number.” In their article, Forbes Contributor Theo Priestley states, “But Klout has now matured beyond this kind of audience now, and as a first step the publication of this paper of how their data scientist team employs sophisticated algorithms and machine learning techniques proves that Klout and Lithium Technologies is much more than just a number.”

 

Since the Klout data team was in the mood to “tell all,” we decided to get even more behind-the-scenes.

We wanted to know how the IEEE BigData audience reacted to the paper. To get our answer, we spoke with Adithya Rao, lead research engineer behind the Klout Score, and asked:

 

What were the top 3 questions and how did you respond to them?

 

  1. Is the Klout Score available in multiple languages?

Adithya: It definitely is. The Klout score is based on actions taken irrespective of language, so – in other words – it’s available in all languages.

 

  1. Does the Klout Score include sentiment?

Adithya: Not at this time. Sentiment is something we have considered adding to the score and might do so in the future. A point to note is that sentiment inference is itself not always accurate, and a negative perception from someone may be deemed as positive by another. So we’re working to see if adding sentiment would improve the accuracy of the Klout Score. More to come on that topic.

 

  1. What networks are scored?

Adithya: Currently, Twitter, Facebook, YouTube, Google+, LinkedIn, Instagram, and Foursquare contribute to the Klout Score as well as Lithium Communities and real world information from Wikipedia.

You can read the entire paper here.

 

There you have it. A lot of data goes into each Klout Score. Have you checked yours out lately? Go to Klout.com to find out your Klout Score. And, not just that, go to Klout.com and find out what you’re an expert in via Klout Topic Expertise. To be considered a Klout Expert on a topic, users must be in the top few percentages of all global users across Klout’s nearly 10,000 topics.

 

Be sure to share your Klout Score and Topic Expertise and any questions for Adithya at @Klout or in the comments section below.

Francisca
Francisca Fanucchi

Francisca is a Klout expert that has worked with Klout since 2014.