Monitoring the count of errors in a time period before rasing notification

We want to be able to monitor events that occour and if a number greater than X occour over Y time then trigger a notification or elevate the events severity. I want to know if this is possible with Zenoss. If so where might the documentation or steps to do it would be found. 

The details:
We have self-healing kiosks. For the most part when they encounter an error they will just reboot resolving 99% of their errors. 
By using the Zenoss's API we push information from these kiosks into Zenoss. However because most of the time they successfully self-heal, we get lots of notifications about errors - that end up being resolved (we do not send clear events).

If this feature is not supported, is there another workaround or sugestion that would assist in this time of management?