Appeal 2007-1246
Application 10/014,180
control value equals the distance between symbolic features and whether
Bieganski discloses aggregating the distances.
Bieganski discloses (col. 9, ll. 24-27 and 47-54) that a serendipity
control value is computed by applying a serendipity control function to each
item in a community popularity data set. The results are values between 0
and 1, with 1 meaning that the item meets the serendipity requirements, and
0 meaning that the item does not meet the serendipity requirements.
Bieganski further teaches (col. 9, l. 64-col. 10, l. 35) that a common control
function includes a high frequency cutoff and a low frequency cutoff. In
other words, the control function would assign a zero to all items with a
community popularity value above and below a specified range. All items
within the range would be assigned a value of one. The serendipity control
values do not indicate overall similarities between items, but rather indicate
weighted community popularities, wherein the weights are not necessarily
applied linearly. In the example of the common control function, the
weights are applied the same to high and low popularities, not increasing or
decreasing with popularity. Therefore, Bieganski's computing serendipity
control values does not equate to computing degrees of similarities, or
distances, between corresponding symbolic features. Further, we find no
disclosure in Bieganski of computing such distances.
Since we have found no disclosure of calculating distances, we
likewise find no disclosure of aggregating distances. Nonetheless, even if
we were to accept the Examiner's position that Bieganski's control values
equal the claimed distances, Bieganski still fails to disclose aggregating the
distances. Bieganski discloses in the portion of column 14 cited by the
Examiner that the serendipity control values are multiplied by the
4
Page: Previous 1 2 3 4 5 6 7 8 9 10 Next
Last modified: September 9, 2013