Currently an effort is underway to allow the public to provide input in the form of prior art submissions fo pending US patent applications (see link). This effort is in response to the increasing number of patent applications (~300,000/year) and an increased difficulty in finding prior art in such fields as software, business methods, and nanotechnology because most of the prior art in these fields are not readily available to patent examiners.
Participation in this program is voluntary on the part of the patent applicant, but those participating may receive incentives such as more expedient processing of their patent applications.
However, given the large number of patent applications submitted to the USPTO, if a sizable percentage of patent applicants decide to participate in the program and there are not enough peer reviewers to provide feedback, it may be difficult to prioritize which applications to review.
In anticipation of this it may be a good idea to provide a patent ranking system in terms of the patent scope wherein broader patent claims (which are more likely to be anticipated by prior art) are pushed to the headof the line of the peer review. One simple way to automate such a procedure would be to write software code to
a) identify the independent claims of new patent applications, and
b) sort the applications beginning with those applications with the fewest number of words in the independent claims.
Another more accurate sorting technique may include
a) identifying the independent claims of new patent applications,
b) identify the words in the independent claims associated with a technology (i.e. transistor, nanotube, glucose, etc.) by comparison with a words in a preestablished database,
c) assign scores to the technology specified words found in b) based on the frequency with which the words are found in the prior art, with higher scores associated with more common technology specified words (i.e. a more common technical term such as transistor would have a higher value than a term such as nanotube), note:the scores would be time-dependent because relative frequency of technical term usage alters with time,
d) establish a total score for the independent claims by adding the scores of all the technology specified words of c)
e) give priority to the applications with the lower scores.
The above technique could focus the peer review to patent applications which most need external review.-- blaise, May 09 2006 Peer to Patent Website http://dotank.nyls.edu/communitypatent/ [blaise, May 09 2006] random, halfbakery