Have Geek Brain

How to create a MEME algorithm?

if you don't know what mem means you can read this readwriteweb article

My question is how to create a meme algorithm, I have a website that brings together thousands of blog posts and I want to share the most talked about stories.

see this quote from the article above

“Aggregating Meme tries to cut the signal-to-noise ratio by figuring out what's most discussed about the news (and hopefully most importantly).

Does anyone know how to do this ?,

are their simple textbooks?

because I'm not that good at math.

thanks

0


a source to share


4 answers


There is no "right" way to do such a thing. There are different ways to achieve this and you need to choose one that you can implement / run and behave however you like. Start with something simple that you understand and work your way up from there.

For instance:

Ben Reeves suggested "(blog numbers with specific tags / total blog posts)" and a pager approach. If these featured topics suit you well, go with them.



Here are some other suggestions,

You can add weight for posts that depend on the popularity of the hosting website. For example, something posted in The New York Times should probably be considered more popular than anything on Joe Schmoos's blog, and should receive more weight. This is similar to the page ranking approach, and in practice it may make little difference.

You can add a time factor, so how quickly posts go on the topic. For instance. If Topic B has 30 posts from last week and Topic C has 10 posts from today, you might want Topic C to become more popular. What if Topic D has 2 posts per week for the last year? What about topic E, which has 5 posts in the last hour?

+1


a source


Variables

  • Count
  • Time
  • Content


Count the number of times the content happens. If it happens often enough, then it qualifies. This must also happen recently, otherwise the score doesn't matter. The content must be well connected in order to avoid false positives.

Take a look at Yahoo contextual search and API keywords for beginners.

+1


a source


Assuming you want to find the most popular topic? The actual calculation can be quite simple, but the amount of data to be processed will be large.

(Number of blog posts with specific blog posts / total volume) = Tag Popularity

Obviously you need a list of common tags / words to ignore

Then the most popular post associated with this tag = Most frequently associated blog post with other posts containing this tag.

Also, more complicated, you can calculate the link weight using the pagerank style calculation. http://www.webworkshop.net/pagerank.html is the likelihood that when randomly browsing, you will land on a specific page i.e. the most popular

/ My 2 cents

0


a source


Get the book "Programming Collective Intelligence" by T. Segaran

http://shop.oreilly.com/product/9780596529321.do

It introduces and explains beautifully the key concepts of popularity and ranking algorithms and provides comprehensive examples in Python.

0


a source






More articles:


All Articles