OneRiot helps you find the news, stories and videos that the social web is buzzing about right now. There is a firehose of realtime information on the social web, and we aim to make sense of that by filtering through the noise and spam to find the most socially-relevant web content related to your search query. We do that with a realtime ranking algorithm we call “Pulse Rank” (think of this as “Page Rank for the realtime web.” More on this to come…). Uniquely, we also rank our results at search time, to give you the most socially relevant results, right now.

Our “Pulse Rank” algorithm looks at dozens of factors that give “weight” to certain results. We made a couple of changes to our algorithm today, to improve those results even further.

Some of the questions we’ve asked ourselves during this process include:

•        Freshness: Is the most recently published content necessarily the most relevant?

•        Domain Authority: Just because I’ve published a post on my own personal blog about Obama, should that be weighted more highly than a post from, say, the New York Times, on the same subject published at the same time?

•        People Authority: Who is sharing this link on the social web? Are they known spammers who pummel their social graph with the same link many times a day, or are they more thoughtful sharers whose links tend to get retweeted and dugg?

•        Acceleration: Is this page increasing in hotness or decreasing in hotness? Are more people sharing the link right now than they were 2 minutes ago? How can you detect an “emerging” webpage vs a popular one that everyone already knows about?

We’re also getting a lot of leverage from our Artificial Intelligence systems that constantly “learn” how to improve the way we rank results.

Here’s a great example:  This is a search on ‘Iran’ on OneRiot, ordered by Pulse Rank. You get the top news that people are sharing right now, the top Youtube video people are watching, and the top opinion out there right now what’s going to happen next.

OneRiot also allows you to see the Firehose of content as it comes into the system. Below is a search on ‘Iran’, but this time it is not ordered by Pulse Rank, but is simply time-based – the FireHose. If it came in most recently, it’s at the top of the list. You can see that the results do include one top news article, but also an old Youtube video about McCain and Iran that someone just republished, and something with only one share (comments on CNN.com). It did come in recently, but it simply doesn’t compare to ordering by Pulse Rank.

Bottomline: with Pulse Rank, the end result to you, the user, should be better results – the most socially relevant content on the web, related to your search query, should be the top result. Give it a spin, and let us know what you think.

Source: OneRiot.com

This entry was posted on Friday, June 19th, 2009 at 9:55 am and is filed under Innovations, Updates. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.