`
`IPR2013-00478 POR at 19-24; ‘352 Jacobs Dec., IPR2013-00478 Exhibit 2113 at ¶ 93-115.
`
`EXHIBIT 2123
`Facebook, Inc. et al. v. Software Rights Archive, LLC
`CASES IPR2013-00478
`IPR2013-00479
`IPR2013-00480
`IPR2013-00481
`
`1
`
`R
`
`
`
`“Petitioners allege Matrix R is the first numerical representation.
`Matrix R however cannot be a first numerical presentation
`because it is not “analyzed” for indirect relationships. Instead,
`Kambayashi teaches that matrix R is merely multiplied by a
`constant (WR) to obtain the Similarity Matrix S. Multiplication
`by a constant cannot be viewed as an “analysis,” much less an
`“analysis for indirect relationships” which require the
`identification of indirect relationships.”
`
`‘352 Jacobs Dec., IPR2013-00478 Exhibit 2113 at ¶ 105; IPR2013-00478 POR at 19-24.
`
`2
`
`
`
`“Furthermore, the Petition conflates Kambayashi’s ID,IDF pairs
`and the R matrix. Based on Petitioners’ error, the Board’s
`Institution Decision suggests the ID,IDF pairs are created from
`Direct Reference Matrix R. However, to be clear, Kambayashi
`does not teach that Matrix R is used to derive ID code pairs,
`and in fact, Matrix R cannot be used to derive the pairs (or the
`bibliographic coupling or co-citation values, either).”
`
`‘352 Jacobs Dec., IPR2013-00478 Exhibit 2113 at ¶ 112; IPR2013-00478 POR at 19-24.
`
`3
`
`
`
`“[I]t is clear that it would not be possible to compute the ID code pairs
`or the bibliographic coupling and co-citation matrices from the matrix
`R. Specifically, the direct matrix R is built from the following values:
`rij = rji = 1 if i (or j) refers to j (or i)
`Since rij = rji in all cases, it is clear that the R matrix is symmetric and thus
`cannot be used to determine any of the following: the ID code pairs, the
`bibliographic coupling values, the B and B’ matrices, the co-citation
`values, or the C and C’ matrices [because it is not possible to tell which
`entry is citing and which entry is cited].”
`
`‘352 Jacobs Decl., IPR2013-00478 Exhibit 2113 at ¶ 110 (citing Kambayashi at 92, internal citations
`omitted); see also IPR2013-00478 POR at 19-24.
`
`4
`
`
`
`Claim 26: “searching the objects in the database using a
`computer and the stored second numerical representations…”
`“Kambayashi does not disclose any searching step, much less a
`searching step “using the stored second numerical
`representations.” The Petition relies on a general statement that
`“clustering is an important tool for efficient retrieval of
`documents in bibliographic database systems.” However,
`“efficient retrieval” can include semantic methods that are
`explicitly excluded from the claim.”
`
`‘352 Jacobs Dec., IPR2013-00478 Exhibit 2113 at ¶ 115; see also IPR2013-00478 POR at 19-24.
`
`5
`
`
`
`“Moreover, this extremely limited disclosure certainly does not
`disclose “using the stored second numerical representation” as
`required by claim 26. Kambayashi does not specifically teach
`searching, and the mere mention of clustering as potentially
`being used for retrieval cannot be seen as explicitly or inherently
`teaching searching the objects in the database using the relied-
`upon, stored, second numerical representations. Clustering has
`non search related uses.”
`
`‘352 Jacobs Dec., IPR2013-00478 Exhibit 2113 at ¶ 115; see also IPR2013-00478 POR at 19-24.
`
`6
`
`
`
`Jacobs Testifies Petitioners Allege Kamabayshi’s ID,IDF
`Codes Are a First Numerical Representation
`Jacobs testifies “Kambayashi does not disclose that ID and IDF are numerical values, and in fact,
`the papers seem to be denoted by identifiers such as ‘EVER7404’ and ‘GARDL7710’. Thus, the
`‘identification codes’ are plainly character strings.”
`
`IPR2013-00478 POR at 19-24;
`‘352 Jacobs Dec., IPR2013-
`00478 Exhibit 2113 at ¶ 100;
`Kambayashi at 3 (Exhibit
`2023)
`
`7
`
`
`
`“Furthermore, Kambayashi consistently refers to the papers or
`the identification “codes” using notations such as ID, IDF, IDi,
`and IDFj. It would be superfluous and inappropriate to refer to
`IDi as the identification code for a paper with identification
`number i because i would then already be the identification code
`for the paper and Idi would refer to the same number. It can be
`further inferred that IDi is not a number since Kambayashi
`clearly discloses references that are not among the numbered
`papers in the similarity matrix.”
`‘352 Jacobs Dec., IPR2013-00478 Exhibit 2113 at ¶ 104; IPR2013-00478 POR at 19-24.
`
`8