Google rolled out the latest Panda update (2.2) on the 16th of June, Increasing the efficiency in which they are able to detect and remove low quality pages from prominent search result positions.
It is important to recognise that Panda is not actually an algorithm update in the traditional sense, but more accurately a new ranking factor which attributes a score to the quality of content on each of your pages. Pages with high quality unique content will attain a preferential score and pass under the filters unhindered, whereas pages with low quality or scraped content will be flagged and penalised. Domains with a high number of flagged pages may even find themselves penalised as a whole.
Panda is based the on the principals of machine learning, a complexed algorithm which allows Google to make intelligent decisions regarding the credibility and value of the content on a web page. Because machine learning algorithms are so complexed, they require an extraordinary amount of computational power to run. For this reason Google does not run currently run Panda continually, but rather on a periodic basis.
The fact that Panda is run periodically means that if your website has been penalised by the update, you cannot expect your rankings to recover until Panda is run again, even if you have effectively resolved the content quality problems which caused you to be penalised in the first place.
There does seem to be a schedule emerging regarding the frequency in which Google run the Panda filter, as since Panda was introduced the filter has been run once each calender month. This schedule has not been to the day however, and the duration between runs has varied from 4 to 7 weeks.
The Panda algorithm is still very much a work in progress, 2.2 is the third update that Google have applied since Panda was originally launched on the 24th of February 2011. Each update is intended to improve the algorithms accuracy and efficiency at detecting low quality content. Matt Cutts (Googles Search quality Spokesperson) announced that the latest update is intended to increase Panda’s efficiency at detecting scraped content in particular, it is therefore possible that a site which displays duplicate content, which was previously unaffected by Panda, may find itself penalised after this, or future updates.