Spam Filter - production

  • 3
  • Question
  • Updated 1 year ago
  • Acknowledged
We are receiving many applications that should have been filtered through the Spam Filter. Several of the applications have the same email address with different biographical information. We've found 5-10 apps with the same email address and then another 5-10 with a different but same email address. 

Would you explain how the filter works or direct me to documentation?

Thank you
Joyce
Photo of Joyce miyabe

Joyce miyabe

  • 1 Post
  • 0 Reply Likes

Posted 2 years ago

  • 3
Photo of Chien Shih

Chien Shih

  • 5 Posts
  • 1 Reply Like
Thanks, Joyce for bringing this up. We also have the same issue, this is a problem especially the student identity entitles to certain privileges.
Photo of Merrie Wales

Merrie Wales, Official Rep

  • 960 Posts
  • 57 Reply Likes
We opened Zendesk tickets to follow up on these issues 
Photo of Merrie Wales

Merrie Wales, Official Rep

  • 960 Posts
  • 57 Reply Likes
All,
For any College that has new SPAM applications please submit a helpdesk request at:  staffsupportccctc@openccc.zendesk.com.  We will request additional information from you about those applications.

Photo of Mitch Leahy

Mitch Leahy

  • 41 Posts
  • 1 Reply Like
We are experiencing a different problem- consistent false positives. One application had a 98% confidence level for being spam. Upon looking at the app we didn't see any problems. Additionally, the student had applied to the college 5 years ago. While we have only encountered 3 so far, it would be great if the metrics for determining spam were more transparent.  It would help us to zero in on potential issues instead of scratching our heads :)
Photo of Merrie Wales

Merrie Wales, Official Rep

  • 960 Posts
  • 57 Reply Likes
Mitch,
For any "false positives" you can mark them as "Valid".  This is a new machine learning tool, and it is learning every day.  Only six Colleges participated in the 'Pilot' training of the tool - so there was limited data to teach the SPAM filter.  We are in the early phase of this tool (less than 30 days since implementation).  Because we do not want the SPAMMER's to learn from this public forum how to successfully submit SPAM applications the specifics of the metrics will not be posted publicly.
Photo of Debra Thayer

Debra Thayer

  • 58 Posts
  • 3 Reply Likes

We have recently received these applications with fake addresses that were not caught by the spam filter.  These are the app_id's:
  

10857094
10740489
10740143
10739157
10739011
10738497
10517395
Photo of Merrie Wales

Merrie Wales, Official Rep

  • 960 Posts
  • 57 Reply Likes
Debra,
You can process these as SPAM applications in the SPAM utility section of the Administrator.  If you need additional assistance please let us know.