Appeal No. 2006-2458
Application No. 10/147,673
probabilities or likelihoods in fig. 4 have a positive value, noting that these
are the values on the “Y axis” [id.]. Appellant further notes that the values
on the horizontal axis represent the difference between the number of
occurrences of a word in the cluster and an expected number of occurrences,
referred to as a significance in Zhilyaev [id.]. Appellant asserts that this
difference (i.e., significance) can be positive or negative [id.]. Appellant
argues that, a negative difference (i.e., significance) is not a negative
likelihood as claimed [brief, page 8, cont’d page 9]. Appellant concludes
that Zhilyaev discloses that all of the likelihoods are positive as shown on
the Y axis of fig 4 [brief, page 9].
The examiner disagrees [answer, page 14, ¶2]. The examiner argues
that Zhilyaev teaches a negative value, at col. 10, line 1, et seq. [id.]. In
response to appellant’s argument that Zhilyaev’s significance value cannot
be equated to a “likelihood,” the examiner notes that Zhilyaev explains how
significance plays a role in determining which word the user intends to enter
(col. 3, lines 1-5) [id.]. The examiner concludes that Zhilyaev explains how
negative values affect the likelihood of a word being a viable option at col.
10, lines 59-65 [id.].
We note at the outset that if the recited term “likelihood” is accorded
its plain, ordinary, and accustomed meaning (i.e., corresponding to a
26
Page: Previous 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Next
Last modified: November 3, 2007