`571.272.7822 Entered:January 30, 2015
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`FACEBOOK, INC., LINKEDIN CORP., and TWITTER, INC.,
`Petitioner,
`
`v.
`
`SOFTWARE RIGHTS ARCHIVE, LLC,
`Patent Owner.
`____________
`
`Case IPR2013-00480
`Patent 5,832,494
`____________
`
`
`Before SALLY C. MEDLEY, CHRISTOPHER L. CRUMBLEY, and
`BARBARA A. PARVIS, Administrative Patent Judges.
`
`PARVIS, Administrative Patent Judge.
`
`
`
`FINAL WRITTEN DECISION
`35 U.S.C. § 318(a) and 37 C.F.R. § 42.73
`
`
`I. BACKGROUND
`
`A.
`
`Introduction
`On July 30, 2013, Facebook, Inc., LinkedIn Corp., and Twitter, Inc.
`(collectively “Petitioner”) filed a Petition (“Pet.”) requesting an inter partes
`review of claims 1, 5, 8, 10, 11, 14–16, 35, and 40 of U.S. Patent No.
`
`
`
`
`
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`5,832,494 (Ex. 1001, “the ’494 Patent”). Paper 2. On February 3, 2014, we
`instituted trial for all challenged claims 1, 5, 8, 10, 11, 14–16, 35, and 40 of
`the ’494 Patent on certain of the grounds of unpatentability alleged in the
`Petition. Paper 17 (“Decision to Institute” or “Inst. Dec.”).
`After institution of trial, Patent Owner, Software Rights Archive, LLC
`(“Patent Owner”), filed a Patent Owner Response (“PO Resp.”). Paper 31.
`Patent Owner also filed a Motion to Amend to cancel claims 8, 10, 11, 35,
`and 40. Paper 32 (“Motion to Amend” or “Mot. Am.”). Patent Owner’s
`Motion to Amend did not propose to add or amend any claims. Mot. Am. 1.
`Petitioner filed a Reply to the Patent Owner Response. Paper 40 (“Reply”).
`A consolidated oral hearing for IPR2013-00478, IPR2013-00479,
`IPR2013-00480, and IPR2013-00481, each involving the same Petitioner
`and the same Patent Owner, was held on October 30, 2014. The transcript of
`the consolidated hearing has been entered into the record. Paper 53 (“Tr.”).
`We have jurisdiction under 35 U.S.C. § 6(c). This final written
`decision is issued pursuant to 35 U.S.C. § 318(a) and 37 C.F.R. § 42.73.
`Petitioner has shown by a preponderance of the evidence that claim 14
`of the ’494 Patent is unpatentable.
`Petitioner has not shown by a preponderance of the evidence that any
`of claims 1, 5, 15, or 16 of the ’494 Patent are unpatentable.
`Patent Owner’s Motion to Amend to cancel claims 8, 10, 11, 35, and
`40 of the ’494 Patent is granted.
`
`B.
`
`Related Proceedings
`Petitioner indicates that the ’494 patent is involved in the following
`co-pending lawsuits: Software Rights Archive, LLC v. Facebook, Inc., No.
`12-cv-3970 (N.D. Cal., filed July 27, 2012), Software Rights Archive, LLC v.
`
`
`
`2
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`LinkedIn Corp., No. 12-cv-3971 (N.D. Cal., filed July 27, 2012), and
`Software Rights Archive, LLC v. Twitter, Inc., No. 12-cv-3972 (N.D. Cal.,
`filed July 27, 2012). Pet. 2. Petitioner also indicates that the ’494 patent
`was the subject of prior litigation: Software Rights Archives, Inc. v. Google,
`No. 08-cv-3172 (N.D. Cal.) (“Google Litigation”). Pet. 9.
`Petitioner filed another petition, IPR2013-00479, which also seeks
`inter partes review of the ’494 patent. The ’494 patent was the subject of
`reexamination no. 90/011,014. Additionally, Petitioner filed other petitions
`on related patents including: (1) IPR2013-00478, which seeks inter partes
`review of U.S. Patent No. 5,544,352 (“the ’352 Patent”) and (2) IPR2013-
`00481, which seeks inter partes review of U.S. Patent No. 6,233,571 (“the
`’571 Patent”). The ’352 Patent issued from the parent of the application that
`issued as the ’494 Patent. The ’571 Patent issued from an application that
`was a divisional of the application that issued as the ’494 Patent.
`
`C.
`
`The ’494 Patent
`The ’494 Patent relates to computerized research on databases.
`Ex. 1001, 1:11–13. The ’494 Patent discloses that it improves search
`methods by indexing data using proximity indexing techniques. Id. at 3:20–
`31. According to the ’494 patent, proximity indexing techniques generate a
`quick-reference of the relations, patterns, and similarities found among the
`data in the database. Id. at 3:28–31.
`Figure 2 of the ’494 Patent illustrates the high-level processing of
`software for computerized searching (Ex. 1001, 8:7–8) and is reproduced
`below:
`
`
`
`3
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`
`
`Figure 2 illustrates high-level processing of three main
`programs for computerized searching.
`As shown in Figure 2 above, software system 60 comprises:
`Proximity Indexing Application Program 62, Computer Search Program for
`Data Represented by Matrices (CSPDM) 66, and Graphical User Interface
`(GUI) Program 70. Ex. 1001, 11:29–36. Processing of software system 60
`begins with Proximity Indexing Application Program 62 indexing a
`database. Id. at 11:46–47. Then, CSPDM 66 searches the indexed database
`and retrieves requested objects. Id. at 11:49–53. CSPDM 66 relays the
`retrieved objects to GUI Program 70 to display on a display. Id. at 11:53–
`55.
`
`Software system 60 runs on a computer system comprising, for
`example, a processor of a personal computer. Ex. 1001, 10:11–15. The
`system comprises a display, which displays information to the user. Id. at
`10:43–44. Exemplary displays include: computer monitors, televisions,
`LCDs, and LEDs. Id. at 10:44–46.
`
`
`
`4
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`
`The processor is connected to a database to be searched. Ex. 1001,
`10:18–20. Data in the database may be represented as a node. Id. at 12:29–
`33. Exemplary nodes include an object or a portion of an object, a document
`or section of a document, and a World Wide Web page. Id. at 12:35–38.
`A cluster link generation algorithm may be used alone, or in
`conjunction with other proximity indexing subroutines, and prior to
`searching. Ex. 1001, 21:30–33. The cluster link generation algorithm may
`generate candidate cluster links (id. at 21:64–66) and then derive actual
`cluster links, which are used to locate nodes for display (id. at 22:1–4).
`Actual cluster links are: “a subset of the candidate cluster links . . . which
`meet a certain criteria.” Id. at 22:1–3.
`
`D.
`
`Illustrative Claims
`The independent claims are 1 and 14. Dependent claim 5 depends
`directly from claim 1. Each of dependent claims 15 and 16 depends, directly
`or indirectly, from claim 14.
`Independent claims 1 and 14 illustrate the claimed subject matter and
`are reproduced below:
`1. A method of analyzing a database with indirect
`relationships, using links and nodes, comprising the steps of:
`selecting a node for analysis;
`generating candidate cluster links for the selected node,
`wherein the step of generating comprises an analysis of one or
`more indirect relationships in the database;
`deriving actual cluster links from the candidate cluster
`links;
`identifying one or more nodes for display; and
`displaying the identity of one or more nodes using the
`actual cluster links.
`
`
`
`5
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`
`14. A method for representing the relationship between
`nodes using stored direct links, paths, and candidate cluster
`links, comprising the steps of:
`a) initializing a set of candidate cluster links;
`b) selecting the destination node of a path as the selected
`node to analyze;
`c) retrieving the set of direct links from the selected node
`to any other node in the database;
`d) determining the weight of the path using the retrieved
`direct links;
`repeating steps b through d for each path; and
`e) storing the determined weights as candidate cluster
`
`links.
`
`E.
`
`
`The Prior Art References Supporting Alleged Unpatentability
`Edward A. Fox, Some Considerations for Implementing the SMART
`Information Retrieval System under UNIX, (Sept. 1983) (Ph.D. dissertation,
`Cornell Univ. Dept. of Comp. Sci.) (“Fox SMART”) (Ex. 1005).
`Edward A. Fox, Extending the Boolean and Vector Space Models of
`Information Retrieval with P-Norm Queries and Multiple Concept Types,
`(Aug. 1983) (Ph.D. dissertation, Cornell Univ. Dept. of Comp. Sci.) (“Fox
`Thesis”) (Ex. 1008).
`The parties do not dispute the prior art status of the references.
`
`The Pending Grounds of Unpatentability
`Reference
`Basis
`Claims challenged
`14–16
`Fox Thesis
`§ 102
`
`§ 102
`
`1 and 5
`
`Fox SMART
`
`
`6
`
`F.
`
`
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`
`II. ANALYSIS
`
`A.
`1.
`
`Claim Construction
`Principles of Law
`Petitioner asserts, and Patent Owner does not dispute, that the ’494
`Patent expired on June 14, 2013. Pet. 9. The Board’s interpretation of the
`claims of an expired patent is similar to that of a district court’s review. See
`In re Rambus, Inc., 694 F.3d 42, 46 (Fed. Cir. 2012). We, therefore, are
`guided by the principle that the words of a claim “are generally given their
`ordinary and customary meaning,” as understood by a person of ordinary
`skill in the art in question at the time of the invention. Phillips v. AWH
`Corp., 415 F.3d 1303, 1312, 13 (Fed. Cir. 2005) (en banc). “In determining
`the meaning of the disputed claim limitation, we look principally to the
`intrinsic evidence of record, examining the claim language itself, the written
`description, and the prosecution history, if in evidence.” DePuy Spine, Inc.
`v. Medtronic Sofamor Danek, Inc., 469 F.3d 1005, 1014 (Fed. Cir. 2006)
`(citing Phillips, 415 F.3d at 1312–17). There is a “heavy presumption,”
`however, that a claim term carries its ordinary and customary meaning. CCS
`Fitness, Inc. v. Brunswick Corp., 288 F.3d 1359, 1366 (Fed. Cir. 2002).
`
`2.
`
`Overview of the Parties’ Positions
`In the Decision to Institute, we construed “cluster links,” “candidate
`cluster links,” “indirect relationships in the database,” “displaying,” and
`“actual cluster links.” Our constructions are set forth in the table below.
`Claim Term or Phrase
`Construction
`“cluster links”
`“[R]elationships used for grouping
`interrelated nodes.” Inst. Dec. 9.
`“[A] set of possible cluster links
`between a search node and a target
`node.” Inst. Dec. 10.
`7
`
`“candidate cluster links”
`
`
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`
`Claim Term or Phrase
`“indirect relationships in the
`database”
`
`“displaying” “[D]epic
`
`“actual cluster links”
`
`Construction
`“[R]elationships that are
`characterized by at least one
`intermediate node between two
`nodes.” Inst. Dec. 11.
`ting information on a
`hardware display.” Inst. Dec. 12.
`“[A] subset of the candidate cluster
`links[,] which meet certain criteria.”
`Inst. Dec. 13.
`
`
`We also determined that “wherein the step of generating comprises an
`analysis of one or more indirect relationships in the database” requires no
`express construction other than the construction of “indirect relationships in
`the database” noted above. Inst. Dec. 11. We further determined that an
`express construction is not necessary for “selecting a node for analysis.”
`Inst. Dec. 11.
`Patent Owner provides contentions based on Patent Owner’s view that
`the claims recite a specific arrangement of steps. PO Resp. 1–2. Patent
`Owner also makes arguments based on a construction of “cluster links” that
`is narrower than our construction. Id. Below, we evaluate whether the
`challenged claims recite a specific arrangement of steps. We also evaluate
`whether to adopt a construction of “cluster links” that is narrower than the
`construction that we adopted in our Decision to Institute.
`With the exception of the arrangement of steps of the challenged
`claims and the construction of “cluster links,” for each of the other claim
`terms above, we discern no reason, based on the complete record now before
`us, to change our construction thereof.
`
`
`
`8
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`3.
`
`Order of Steps
`Patent Owner contends that the challenged claims of the ’494 Patent
`are arranged in a specific manner. PO Resp. 2. For example, Patent Owner
`contends that claim 1 requires a specific arrangement in which “deriving
`actual cluster links from the candidate cluster links” is performed after
`“generating candidate cluster links for the selected node.” Id. at 16. Patent
`Owner similarly contends that claim 14, from which claim 15 depends,
`requires a specific arrangement. Id. at 48 (citing Ex. 2113 ¶ 205). Patent
`Owner contends that the prior art does not disclose the specific arrangement
`of the challenged claims.
`Petitioner does not take a position on whether the elements of the
`challenged claims require the specific arrangement argued by Patent Owner.
`Petitioner contends that the asserted prior art discloses the steps as arranged
`in the challenged claims. See e.g., Reply 9.
`To evaluate Patent Owner’s and Petitioner’s contentions, we
`determine whether claims 1 and 15 recite a specific arrangement of steps
`such that deriving actual cluster links is performed after candidate cluster
`links are generated, as recite in claim 1, or after a set of candidate cluster
`links is initialized and stored, as recited in claim 15.
`The full recitations of the phrases of claim 1 at issue are below.
`1. A m ethod of analyzing a da tabase . . . co mprising the
`steps of:
`. . .generating candidate cluster links for the selected
`node, wherein the step of generating comprises an analysis of
`one or more indirect relationships in the database;
`deriving actual cluster links from the candidate cluster
`links. . . .
`
`
`
`9
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`(Emphases added). Claim 15 similarly recites deriving the actual cluster
`links after the steps of initializing and storing candidate cluster links.
`We determine that actual cluster links can be derived from candidate
`cluster links only if the candidate cluster links already have been generated.
`This determination is consistent with the specification of the ’494 Patent. In
`particular, the specification of the ’494 Patent explains that candidate cluster
`links are the set of all possible cluster links between a search node and a
`target node. Ex. 1001, 21:64–22:1. The ’494 Patent specification continues
`that actual cluster links are a subset of the candidate cluster links, which
`meet certain criteria. Id. at 22:1–4.
`We agree with Patent Owner that claims 1 and 15 recite a specific
`arrangement with respect to the two steps of generating candidate cluster
`links and deriving actual cluster links from the candidate cluster links, as
`recited in claim 1, and as commensurately recited in claim 15. In particular,
`actual cluster links are derived from candidate cluster links after the
`candidate cluster links have been generated, as recited in claim 1, or stored,
`as recited in claim 15.
`
`4.
`
`“cluster links”
`The term “cluster links” is recited, for example, in claims 1 and 14,
`within “candidate cluster links.” The term “cluster links” also is recited in
`claims 1 and 15 within “actual cluster links.” In the Decision to Institute, we
`construed “cluster links” in light of the specification to mean “relationships
`used for grouping interrelated nodes.” Inst. Dec. 9.
`Patent Owner contends that the asserted art describes
`“experimentation with relationships existing among printed documents.” PO
`Resp. 1. Patent Owner further contends that the challenged claims of the
`
`
`
`10
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`’494 Patent are directed to analyzing and searching a computer database of
`objects. Id. Petitioner contends that Patent Owner’s argument that the prior
`art describes relationships between paper documents, not electronic ones, is
`irrelevant because nothing in the challenged claims requires a database of
`objects that cite to other objects. Reply 1.
`To evaluate Patent Owner’s and Petitioner’s contentions, we
`determine whether “cluster links” means relationships between two nodes,
`which are represented in data stored in a computer. Claim 1 recites a
`method of analyzing data in a database with indirect relationships using links
`and nodes. Claim 14 similarly recites a method for representing the
`relationship between nodes, using stored direct links, paths, and candidate
`cluster links.
`As explained in the ’494 Patent specification, data in the database may
`be represented as a node. Ex. 1001, 12:29–33. Exemplary nodes include an
`object or a portion of an object, a document or section of a document, and a
`World Wide Web page. Id. at 12:35–38. The ’494 Patent specification
`states that a link is a “relationship between two nodes.” Id. at 12:65–66.
`The ’494 Patent specification continues, “[a] link [] can be represented by a
`vector or an entry on a table and contain information for example, a from-
`node identification [] (ID), a to-node ID [], a link type [], and a weight.” Id.
`at 13:3–6. As described in the ’494 Patent specification, a link is
`represented in data stored in a computer.
`We, therefore, determine that “cluster links” means relationships,
`which are represented in data stored in a computer and are used for grouping
`interrelated nodes.
`
`
`
`11
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`B.
`
`Alleged Anticipation of Claims 1 and 5 by Fox SMART
`Petitioner contends that claims 1 and 5 of the ’494 Patent are
`unpatentable, under 35 U.S.C. § 102, as anticipated by Fox SMART.
`Pet. 17–20. To establish anticipation, each and every element in a claim,
`arranged as is recited in the claim, must be found in a single prior art
`reference. Net MoneyIN, Inc. v. VeriSign, Inc., 545 F.3d 1359, 1369 (Fed.
`Cir. 2008); Karsten Mfg. Corp. v. Cleveland Golf Co., 242 F.3d 1376, 1383
`(Fed. Cir. 2001). We determine that Petitioner has not shown by a
`preponderance of the evidence that claims 1 and 5 are unpatentable as
`anticipated by Fox SMART.
`
`1.
`
`Fox SMART
`The System for Mechanical Analysis and Retrieval of Text (SMART)
`is described as a project for designing a fully automatic document retrieval
`system and for testing new ideas in information science. Ex. 1005, 3.
`In SMART, an automatic indexing component constructs stored
`representations of documents. Ex. 1005, 3. Bibliographic information is
`used to enhance document representations. Id. at 29. The SMART system
`may process basic raw data, such as an exemplary N collection of articles
`and citation data describing which articles are cited by others. Id. at 29–30.
`The exemplary input data includes indirect citation relationships, such as
`bibliographic coupled and co-citation relationships. Id. at 30–32.
`A clustering algorithm is processed by the SMART system as follows:
`“[t]he clustering algorithm produces a hierarchy where all N documents in a
`collection end up as leaves of a multilevel tree. . . . Clustering proceeds by
`adding documents one by one starting with an initially empty tree.” Ex.
`1005, 44. Adding documents involves finding the proper place to insert,
`
`
`
`12
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`attaching the incoming entry appropriately, and recursively splitting overly
`large nodes. Id. at 47.
`
`In addition to splitting, the SMART system may delete clusters that
`exhibit too much overlap and assign others to a garbage or orphan cluster.
`Ex. 1005, 49. Eventually, only clusters that pass all appropriate tests are
`accepted. Id. at 51.
`
`2.
`
`Claims 1 and 5
`Petitioner points to Fox SMART’s description of a clustering
`algorithm as disclosing generating candidate cluster links for a selected node
`by analyzing indirect relationships in a database, as recited in claim 1. Pet.
`18–19 (citing Ex. 1005, 30–32, 44, 46; Ex. 1009 ¶¶ 160–61). In particular,
`Petitioner points to Fox SMART’s description of constructing bibliographic
`and co-citation subvectors by analyzing indirect relationships, specifically
`bibliographic and co-citation relationships. Id. at 18 (citing Ex. 1005, 30–
`32). According to Petitioner’s Declarant, Dr. Edward A. Fox, the resulting
`bibliographic and co-citation subvectors are stored in a computer for further
`computations using the clustering procedure. Ex. 1009 ¶¶ 173–74 (citing
`Ex. 1005, 36–38, 46). Petitioner additionally points to Fox SMART’s
`description that the clustering algorithm produces a hierarchy or
`classification in which “all” of the N documents in the collection end up as
`leaves of a multilevel tree. Pet. 18–19 (citing Ex. 1005, 44). The resulting
`tree, according to Petitioner, is a set of possible cluster links between a
`search node and a target node. Id.; see also Ex. 1009 ¶ 160 (“Fox SMART .
`. . disclose[s] generating candidate cluster links . . . For example, the
`particular clustering analysis that I employed builds a tree.”).
`
`
`
`13
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`
`Petitioner, however, points to this same clustering algorithm for
`deriving actual cluster links from the candidate cluster links, as recited in
`claim 1. Pet. 19 (citing Ex. 1005, 47, 49–51; Ex. 1009 ¶ 162). In particular,
`Petitioner points to Fox SMART’s description of concentration tests
`performed as part of the clustering algorithm. Id. (citing Ex. 1005, 51). As
`described in Fox SMART:
`Candidate clusters which pass the concentration test are those
`formed by having enough highly correlated pairs in the
`proposed cluster. . . .
` “Uncour” repeatedly considers the remaining cluster that
`is most heavily covered by other clusters. If the overlap is too
`much, it is deleted. Eventually only clusters that pass all
`appropriate tests are accepted.
`Ex. 1005, 50–51.
`Petitioner’s Declarant, Dr. Fox, testifies that Fox SMART derives a
`subset because “clusters that do not pass all the concentration and overlap
`tests are deleted.” Ex. 1009 ¶ 162 (citing Ex. 1005, 51). Dr. Fox
`supplements his testimony by stating, “Fox SMART teaches that potential
`(i.e., candidate) clusters are rejected (‘deleted’) if they fail any one of these
`tests.” Ex. 1028 ¶ 281 (citing Ex. 1005, 51). Dr. Fox further states that
`claim 1 does not require that candidate cluster links be deleted. Id.
`Patent Owner contends that Dr. Fox incorrectly states that clusters that
`do not pass all of the tests are deleted. PO Resp. 33 (citing Ex. 2113 ¶¶ 82–
`103). Patent Owner’s Declarant, Dr. Paul S. Jacobs, provides his analysis of
`various aspects of the clustering process and concludes that the clustering
`algorithm, including the concentration tests noted above, does not result in a
`subset. Ex. 2113 ¶¶ 82–103. In particular, Dr. Jacobs states that the
`clustering process involves accepting trial clusters, which then must pass the
`
`
`
`14
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`concentration tests to become candidate clusters. Id. ¶ 91. Dr. Jacobs also
`states that clusters are deleted only in the case of overlap with a new group
`of clusters formed from splitting. Id. ¶ 88. Additionally, Dr. Jacobs states
`that moving orphans to the garbage cluster does not result in deleting those
`orphans or creating a subset. Id. ¶ 101. Dr. Jacobs, instead, states that the
`orphans may not be garbage in the end as they may be assigned a node as
`new documents are added to the tree. Id.
`We determine that the statements of Patent Owner’s Declarant are
`consistent with Fox SMART’s description of clustering. Fox SMART
`describes the clustering process as initializing a new tree as empty, adding
`documents to the tree, and recursively splitting overly large nodes of the
`tree. Ex. 1005, 47. Fox SMART states that splitting is accomplished by the
`following procedures: div_cent, cleave, and uncour. Id. at 49. Fox SMART
`further describes the splitting process as follows:
`First a complete similarity matrix is formed based on the
`pairwise combined similarity values. “Cleave” then identifies a
`plausible clustering except that no limit on overlap is
`considered. “Uncour” compensates for that by first deleting
`clusters that exhibit too much overlap with remaining clusters,
`and secondly by assigning the others to a “garbage” or “orphan”
`cluster.
`
`Id.
`
`As described in Fox SMART, the concentration tests that are cited by
`Petitioner are performed as part of forming the cluster tree. Id. Petitioner
`does not identify disclosure in Fox SMART of deleting clusters other than
`those that simply overlap, or duplicate, other clusters. Overlapping clusters
`are deleted following a routine that identifies plausible clustering with “no
`limit on overlap.” Id. Additionally, Fox SMART includes code that collects
`
`
`
`15
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`orphans. Id. at 52. Petitioner, however, has not shown that Fox SMART
`describes a subset which does not include these orphans.
`Dr. Fox’s supplemental testimony that “Fox SMART teaches that
`potential (i.e., candidate) clusters are rejected (‘deleted’) if they fail any one
`of these tests” (Ex. 1028 ¶ 281 (citing Ex. 1005, 51)) suggests that the terms
`“rejected” and “deleted” are the same. We do not agree. In view of Dr.
`Fox’s testimony in both of his Declarations and the testimony of Dr. Jacobs,
`we find that one of ordinary skill in the art reasonably would have
`understood Fox SMART as describing deleting overlap that had been
`generated by the immediately preceding software routine. Additionally, we
`find that one of ordinary skill in the art reasonably would have understood
`Fox SMART as describing that the tests referred to by Dr. Fox are processed
`during formation of the tree. We, therefore, determine that Dr. Fox’s
`testimony does not address persuasively the requirement in claim 1 of
`deriving a subset of the already generated candidate cluster links.
`Claim 1 additionally recites “displaying the identity of one or more
`nodes using the actual cluster links.” For this element, Petitioner points to
`Fox SMART’s description of searching a clustered tree. Pet. 19 (citing Ex.
`1005, 53–54; Ex. 1009 ¶ 163). Fox SMART states that “one would like to
`retrieve and rank documents so that all relevant documents, regardless of
`what cluster they appear in, are retrieved as soon as possible.” Ex. 1005, 53.
`Fox SMART continues that “most of the documents in a retrieved cluster are
`presented to the user.” Id. at 54.
`We are not persuaded that Fox SMART’s description of ranking
`documents discloses deriving a subset because a set of ranked documents
`provides an indication of an order of presentation, but is not a subset.
`
`
`
`16
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`Additionally, Fox SMART indicates that documents from multiple clusters
`are ranked. Furthermore, Petitioner does not point to disclosure in Fox
`SMART of criteria for forming a subset. Because Petitioner does not point
`to disclosure of deriving a subset, Petitioner has not shown by a
`preponderance of the evidence display using links of that subset.
`In light of the Declaration by Patent Owner’s Declarant, Dr. Jacobs,
`we determine that Petitioner has not shown by a preponderance of the
`evidence that Fox SMART discloses either: (1) deriving actual cluster links
`for the candidate cluster links; or (2) displaying the identity of one or more
`nodes using the actual cluster links. For the foregoing reasons, Petitioner
`has not established by a preponderance of the evidence that claim 1 is
`anticipated by Fox SMART. Because claim 5 depends from claim 1, we
`also determine that Petitioner has not established by a preponderance of the
`evidence that claim 5 is anticipated by Fox SMART.
`
`C.
`
`Alleged Anticipation of Claims 14–16 by Fox Thesis
`Petitioner contends that claims 14–16 of the ’494 Patent are
`unpatentable, under 35 U.S.C. § 102, as anticipated by Fox Thesis. Pet. 10–
`16. In support of the asserted ground of unpatentability, Petitioner sets forth
`the disclosure of Fox Thesis, provides a detailed claim chart, and cites to the
`declaration of Dr. Fox (Ex. 1009 ¶¶ 156–78), explaining how each limitation
`is disclosed in Fox Thesis. Pet. 10–16.
`Petitioner’s claim chart persuasively reads all elements of claim 14
`onto the disclosure of Fox Thesis. Despite the counter-arguments in Patent
`Owner’s Response, and the evidence cited therein, which we have also
`considered, Petitioner has shown by a preponderance of the evidence that
`claim 14 is unpatentable as anticipated by Fox Thesis.
`
`
`
`17
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`
`We, however, determine that Petitioner has not shown by a
`preponderance of the evidence that claims 15 and 16 are unpatentable as
`anticipated by Fox Thesis.
`
`1. Fox Thesis
`Fox Thesis describes improving query and document representation
`schemes for information retrieval. Ex. 1008, 261. In particular, useful types
`of bibliographic data are incorporated into a model to test clustering and
`retrieval functions. Id. at 164.
`Bibliographic connections between articles are illustrated for an
`exemplary set “O” of documents, which are represented by letters A through
`G. Ex. 1008, 165, 66; Fig. 6.2. This exemplary set “O” includes direct and
`indirect citation references. Id. at 166, 67; Table 6.2.
`Based on the reference pattern for a set of documents, Fox Thesis
`describes deriving various measures of the interconnection between the
`documents. Ex. 1008, 166. For example, weights are assigned “based upon
`integer counts” for bibliographically coupled documents. Id. at 167.
`Citation submatrices represent reference or citation information.
`
`Ex. 1008, 169. For example, submatrix (cid:1854)(cid:1855) represents bibliographically
`coupled reference information and submatrix (cid:1855)(cid:1855) represents co-citation
`
`reference information. Id. at 169–72; Figs. 6.3–6.5.
`
`2.
`
`Claim 14
`Petitioner’s claim chart persuasively reads all elements of claim 14
`onto the disclosure of Fox Thesis. Pet. 14–16 (citing Ex. 1008, 164–68,
`170–72, 174–77, 181–82, 193, 195, 213, 237–39, 261, 272; Ex. 1009
`¶¶ 157–61, 169, 172–74). We address Patent Owner’s counter-arguments in
`turn.
`
`
`
`18
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`
`Patent Owner contends that Petitioner “slap[s] together” disparate
`quotes from unrelated portions of Fox Thesis. PO Resp. 3. For example,
`Patent Owner contends that a user query relied on by Petitioner for
`disclosing selecting a destination node, as recited in claim 14, does not relate
`“in any way” to steps of a clustering process identified by Petitioner for
`other steps of claim 14. PO Resp. 47–48.
`Fox Thesis, however, describes interrelated experiments. Ex. 1008,
`21. In particular, Fox Thesis describes that the user query relied on by
`Petitioner (Pet. 12 (citing Ex. 1008, 237–39)) is used “[t]o quickly test the
`utility of extended vector feedback” (Ex. 1008, 237). As is evident in Fox
`Thesis, a query may be used to initiate a search. Id. at 234. Additionally,
`according to Fox Thesis, searches are conducted using clustering processes.
`Id. at 218. We, therefore, are persuaded that Petitioner shows by a
`preponderance of the evidence that Fox Thesis describes selecting a
`destination node, as recited in claim 14.
`Patent Owner additionally contends that Fox Thesis does not disclose
`retrieving the set of direct links, as recited in claim 14. PO Resp. 49. In
`particular, Patent Owner contends that although the Fox Thesis describes
`
`using an (cid:1864)(cid:1866) sub-vector as a representation of direct links, the (cid:1864)(cid:1866) links are not
`
`used for obtaining the sub-vectors representing the indirect links. Id. Patent
`Owner further argues that retrieving the set of direct links is not a necessary
`precursor to finding co-citation values. Id. at 50.
`Petitioner, however, points to Fox Thesis’s description of a reference
`pattern for a set of documents and measures of interconnection between
`those documents. Pet. 15 (citing Ex. 1008, 166–168). As set forth in Fox
`Thesis, this discussion pertains to direct references between documents, as
`
`
`
`19
`
`
`
`IPR2013-00480
`Patent 5,832,494
`
`well as indirect references. Ex. 1008, 167. Additionally, Dr. Fox testifies
`that retrieving direct links is a precursor to finding indirect relationships.
`Ex. 1009 ¶ 171. Dr. Fox’s testimony is consistent with Fox Thesis’s
`description of a distance “k,” which designates the number of “arcs”
`between a document and another cited document. Ex. 1008, 167. We,
`therefore, are persuaded that Petitioner’s identification of the above
`disclosure shows by a preponderance of the evidence that Fox Thesis
`describes retrieving the set of direct links, as recited in claim 14.
`Patent Owner, in reliance on its Declarant, Dr. Jacobs, further
`contends that co-citation count