Comments on: Sphinn spam – some solutions http://www.scienceforseo.com/social-networks/sphinn-spam-some-solutions/ a bridge between worlds Sat, 05 Feb 2011 14:02:08 +0000 hourly 1 By: andymurd http://www.scienceforseo.com/social-networks/sphinn-spam-some-solutions/comment-page-1/#comment-20 andymurd Thu, 13 Nov 2008 17:34:00 +0000 http://www.scienceforseo.com/wordpress/?p=109#comment-20 I didn't use an SVM but I should definitely look into it. <br/><br/>I tried to get around the lack of a spam feed by classifying posts by topic and grouping topics into "good" and "bad". A lot of Sphinn's spam submissions are well-written web pages, but they're just not relevant, like posts about holidays to India.<br/><br/>It was a fun exercise and I'll pick it up again.<!-- Touched by JuLiA --> I didn’t use an SVM but I should definitely look into it.

I tried to get around the lack of a spam feed by classifying posts by topic and grouping topics into “good” and “bad”. A lot of Sphinn’s spam submissions are well-written web pages, but they’re just not relevant, like posts about holidays to India.

It was a fun exercise and I’ll pick it up again.

]]> By: CJ http://www.scienceforseo.com/social-networks/sphinn-spam-some-solutions/comment-page-1/#comment-19 CJ Thu, 13 Nov 2008 17:25:00 +0000 http://www.scienceforseo.com/wordpress/?p=109#comment-19 I think this is an in-house issue, because they have all of the data, and without it you can't reliably train and test a classifier. <br/><br/>Have you tried using an SVM with the Naive Bayes? <br/><br/>Def write something up and have another go :)<!-- Touched by JuLiA --> I think this is an in-house issue, because they have all of the data, and without it you can’t reliably train and test a classifier.

Have you tried using an SVM with the Naive Bayes?

Def write something up and have another go :)

]]> By: andymurd http://www.scienceforseo.com/social-networks/sphinn-spam-some-solutions/comment-page-1/#comment-18 andymurd Thu, 13 Nov 2008 17:08:00 +0000 http://www.scienceforseo.com/wordpress/?p=109#comment-18 I tried training a bayesian classifier to determine sphinn spam a couple of weeks ago. I got a reasonable start - 90% of spam correctly identified, 10% false negative, 1-3% false positive.<br/><br/>These figures are good but not good enough for a production website. I couldn't improve them further due to a couple of problems:<br/><br/>1. There is no publicy available sphinn spam data. Grabbing the upcoming and hot feeds and assuming anything that gets dumped within 3 hours is not accurate enough. I'd need an RSS feed of known spam.<br/><br/>2. Sphinn has a surprisingly wide variety of posts. You'd think that stories about gambling should be off topic for Sphinn, but no, <a HREF="http://sphinn.com/story/79400" REL="nofollow">this went hot</a>.<br/><br/>There are a lot of other good factors that could help, such as the age of the poster's account, number of repeated words in the comment, URL depth etc.<br/><br/>I guess I should have another go, or at least write up my findings.<!-- Touched by JuLiA --> I tried training a bayesian classifier to determine sphinn spam a couple of weeks ago. I got a reasonable start – 90% of spam correctly identified, 10% false negative, 1-3% false positive.

These figures are good but not good enough for a production website. I couldn’t improve them further due to a couple of problems:

1. There is no publicy available sphinn spam data. Grabbing the upcoming and hot feeds and assuming anything that gets dumped within 3 hours is not accurate enough. I’d need an RSS feed of known spam.

2. Sphinn has a surprisingly wide variety of posts. You’d think that stories about gambling should be off topic for Sphinn, but no, this went hot.

There are a lot of other good factors that could help, such as the age of the poster’s account, number of repeated words in the comment, URL depth etc.

I guess I should have another go, or at least write up my findings.

]]>