I/P Engine, Inc. v. AOL, Inc. et al
Filing
427
Opposition to 237 MOTION for Summary Judgment Defendants AOL Inc., Google Inc., IAC Search & Media, Inc., Gannett Company, Inc., and Target Corporation's Motion for Summary Judgment filed by I/P Engine, Inc.. (Attachments: # 1 Exhibit 1, # 2 Exhibit 2-8 (Filed Under Seal), # 3 Exhibit 9, # 4 Exhibit 10, # 5 Exhibit 11-19 (Filed Under Seal), # 6 Exhibit 20, # 7 Exhibit 21, # 8 Exhibit 22-24 (Filed Under Seal), # 9 Exhibit 25, # 10 Exhibit 26, # 11 Exhibit 27-32 (Filed Under Seal), # 12 Exhibit 33, # 13 Exhibit 34-39 (Filed Under Seal), # 14 Exhibit 40, # 15 Exhibit 41, # 16 Exhibit 42, # 17 Exhibit 43, # 18 Exhibit 44, # 19 Exhibit 45 (Filed Under Seal), # 20 Exhibit 46, # 21 Exhibit 47, # 22 Exhibit 48, # 23 Exhibit 49, # 24 Exhibit 50, # 25 Exhibit 51, # 26 Exhibit 52, # 27 Exhibit 53-54 (Filed Under Seal), # 28 Exhibit 55, # 29 Exhibit 56 (Filed Under Seal))(Sherwood, Jeffrey)
Exhibit 25
UNITED STATES DISTRICT COURT
EASTERN DISTRICT OF VIRGINIA
NORFOLK DIVISION
__________________________________________
)
I/P ENGINE, INC.,
)
)
Plaintiff,
)
v.
)
Civ. Action No. 2:11-cv-512
)
AOL, INC. et al.,
)
)
Defendants.
)
__________________________________________)
REBUTTAL EXPERT REPORT OF DR. JAIME
CARBONELL REGARDING VALIDITY OF
U.S. PATENT NOS. 6,314,420 and 6,775,664
75.
With respect to claim 28 of the ‘664 patent, although Dr. Ungar states that
Lashkari anticipates claim 28 of the ‘664 patent, I disagree. For the reasons already noted in
connection with claim 26, from which claim 28 depends, claim 28 is not anticipated by Lashkari.
76.
With respect to claim 38 of the ‘664 patent, although Dr. Ungar states that
Lashkari anticipates claim 38 of the ‘664 patent, I disagree. For the reasons already noted in
connection with claim 26, from which claim 38 depends, claim 38 is not anticipated by Lashkari.
XI.
U.S. PATENT NO. 6,185,558 TO BOWMAN (“BOWMAN”) DOES NOT
INVALIDATE THE PATENTS-IN-SUIT
77.
Defendants and Dr. Ungar assert that Bowman anticipates claims 10, 14, 15, 25,
27 and 28 of the ‘420 patent and claims 1, 5, 6, 21, 22, 26, 28 and 38 of the ‘664 patent. I
disagree.
78.
With respect to claims 10, 14, 15, 25, 27 and 28 of the ‘420 patent, Bowman does
not disclose:
“filtering the informons on the basis of applicable content profile data for relevance
to the query”; and
“combining pertaining feedback data . . . with the content profile data in filtering each
informon for relevance to the query.”
79.
Because Bowman does not disclose all of the limitations of claims 10 and 25 of
the ‘420 patent, Bowman cannot disclose all of the limitations recited by claims 14, 15, 27, and
28 of the ‘420 patent.
80.
With respect to claims 1, 5, 6, 21, 22, 26, 28 and 38 of the ‘664 patent, Bowman
does not disclose:
“searching for information relevant to a query associated with a first user”
“combining the information from the feedback system with the information from the
scanning system and . . . filtering the combined information for relevance to at least
17
one of the query and the first user” or “combining the information found to be
relevant to the query by other users with the searched information [and] . . . filtering
the combined information for relevance to at least one of the query and the first user”
as recited in claims 1 and 26 of the ‘664 patent, respectively.
81.
Because Bowman does not disclose all of the limitations of claims 1 and 26 of the
‘664 patent, Bowman cannot disclose all of the limitations recited by claims 5, 6, 21, 22, 28 and
38 of the 664 patent.
82.
First, Bowman does not disclose “filtering the informons on the basis of
applicable content profile data for relevance to the query” or “searching for information relevant
to a query associated with a first user.” Dr. Ungar states in paragraph 57, as part of his general
description of Bowman, that “search results whose content contains all the terms in the query get
higher ranking scores while search results get progressively lower ranking scores as their content
contains fewer and fewer of the terms in the query.” I disagree. Bowman discloses ranking
items “in accordance with collective and individual user behavior, rather than in accordance with
attributes of the items.” Bowman 2:63-3:2; 4:38-48.2
2
Bowman initially establishes a rating table without any entries. See FIGS. 3, 4 or 6; 5:48-49.
To populate the ranking table with entries, Bowman discloses “the facility identifies all of the
query result item selections made by users during the period of time for which rating table is
being generated.” Bowman 5:49-51. Bowman then “identifies the terms used in the query that
produced the query result in which the item selection took place.” Bowman 5:60-62. The scores
in the rating table are augmented when a user clicks on a result by incrementing the score of each
term in the query, whether or not that term is actually is contained in the search results. Bowman
6:9-25. Bowman’s “facility uses rating tables that it has generated to generate ranking values for
items in new query results.” Bowman 9:28-29. The facility determines the rating score
contained by the most recently generated rating table for the current term and item. Bowman
9:35-39. Bowman’s “facility combines the scores for the current item to generate a ranking
value for the item. Bowman 9:41-43. To process a new query at run time, Bowman consults the
rating table and adds the scores for each terms associated with the item in the query, ranking the
items by their ranking scores. Bowman 9:48-49. Bowman also discloses that “scores may be
adjusted to more directly reflect the number of query terms that are matched by the item.”
Bowman 9:50-53.
18
83.
Dr. Ungar then relies upon a described example using an example query “Paris
museum vacations” to support his point. Dr. Ungar’s example, however, presupposes that
Bowman has full search engine capabilities in combination with the use of a popularity ranking
system when in fact Bowman only discloses the latter. Bowman 2:19-35; 2:63-3:2; 4:38-48.
84.
In paragraph 60 of his report, Dr. Ungar states that “in Bowman, the final ranking
score for each item is generated through “a combination [that includes] . . . content-based
filtering (analyzing the item’s content to see how many of the words from the query appear in the
item).” I disagree. As stated above, Bowman does not analyze the items’ content, but instead
relies on his rating table.3
85.
Specifically referencing “filtering the informons on the basis of applicable content
profile data for relevance to the query” limitation of claim 10 of the ‘420 patent, Dr. Ungar states
in paragraph 105 that Bowman discloses this limitation because “Bowman examines each search
result’s content profile to see how many of the query terms are contained therein.” He goes on to
state that “Bowman adjusts the search results ranking scores by giving higher scores to search
results every term in the query.” First, Dr. Ungar misquotes Bowman. Bowman states “scores
may be adjusted to more directly reflect the number of query terms that are matched by the item,
so that the items that match more query terms are favored in the ranking.” Bowman 4:34-48. As
such, Bowman makes no reference to examining a “search results’ content profile” or any
3
In paragraph 99 of his report, Dr. Ungar states that “Bowman further adjusts the ranking score
of the search results according to how many of the search terms in the query are matched by each
search result.” It is apparent that, by “matched,” Dr. Ungar means that a term of the query is
actually found in the document. In my opinion a person of ordinary skill in the art would
understand that Bowman uses the word “matched” to indicate that term-item association is
contained in Bowman’s rating table rather than contained in the item (document). This opinion
is fully consistent with Bowman’s disclosure of ranking items “in accordance with collective and
individual user behavior, rather than in accordance with attributes of the items.” Bowman 2:593:22; 4:38-48.
19
refers to only user feedback from prior users with no reference to item (document) contents or
other search engine rankings.
88.
In paragraph 114 of his report, Dr. Ungar further states that Bowman discloses the
“‘content-based’ aspect of this claim element.” I again disagree. Dr. Ungar misconstrues
Bowman. Bowman discloses that “scores may be adjusted to more directly reflect the number of
query terms that are matched by an item.” As explained above, in my opinion a person of
ordinary skill in the art would understand that Bowman uses the word “matched” to indicate that
term-item association is contained in Bowman’s rating table rather than contained in the item
(document). This opinion is fully consistent with Bowman’s disclosure of ranking items “in
accordance with collective and individual user behavior, rather than in accordance with attributes
of the items.” Bowman 2:59-3:22; 4:38-48. Bowman does not refer to the search result’s
content, or terms appearing in the search result.
89.
Second, because Bowman does not disclose a content-based information
component, it does not disclose “combining pertaining feedback data . . . with the content profile
data”; “combining the information from the feedback system with the information from the
scanning system”; or “combining the information found to be relevant to the query by other users
with the searched information.”
90.
Third, Bowman does not disclose “filtering each informon on the basis of
applicable content profile data for relevance to the query,” “filtering each informon for relevance
to the query” or “filtering the combined information for relevance to at least one of the query and
the first user” because Bowman does not disclose filtering. Dr. Ungar states that “Bowman then
filters out (i.e., excludes) search results whose ranking scores fall below a certain threshold, or
presents a predetermined number of search results that have the highest ranking scores and filter
21
out all the rest.” First, filtering is different from “subsetting,” the actual technique disclosed in
Bowman on which Dr. Ungar relies. “Subsetting” as disclosed in Bowman is retaining a subset
of a ranked list either by thresholding on ranking values or retaining the top “N” results.
Bowman 9:58-64. These techniques are relative and carried out with reference to the entire
ranked list of search results. The use of these techniques is different than filtering, which does
not use a ranked list, but rather is an item-by-item process.4
91.
Moreover, also for the reasons noted above, Bowman does not disclose filtering.
See Bowman 9:58-67.
92.
With respect to claims 14 and 15 of the ‘420 Patent, although Dr. Ungar states
that Bowman anticipates claims 14 and 15 of the ‘420 patent, I disagree. For the reasons already
noted in connection with claim 10, from which claims 14 and 15 depend, claims 14 and 15 are
not anticipated by Bowman.
93.
With respect to claim 25 of the ‘420 patent, Dr. Ungar states that Bowman
anticipates claim 25 for the same reasons that it anticipates claim 10. Thus, for at least the
reasons that I have opined with respect to claim 10, claim 25 is not anticipated by Bowman.
With respect to claims 27 and 28 of the ‘420 Patent, although Dr. Ungar states that Bowman
anticipates claim 25 of the ‘420 patent, I disagree. For the reasons already noted in connection
with claim 25, from which claims 27 and 28 depend, claims 27 and 28 are not anticipated by
Bowman.
94.
With respect to claim 5 of the ‘664 patent, although Dr. Ungar states that
Bowman anticipates claim 5 of the ‘664 patent, I disagree. For the reasons already noted in
connection with claim 1, from which claim 5 depends, claim 5 is not anticipated by Bowman.
4
In footnote 16, Dr. Ungar states that the “content-based filter system” limitation is obvious
over Bowman in view of Rose. I address this argument in the obviousness section of my report.
22
the combined information for relevance to at least one of the query and the first user”
as recited in claims 1 and 26 of the ‘664 patent, respectively.
105.
Because Culliss does not disclose all of the limitations of claims 1 and 26 of the
‘664 patent, Culliss cannot disclose all of the limitations recited by claims 5, 6, 21, 22, 28 and 38
of the 664 patent.
106.
First, Culliss does not disclose “filtering the informons on the basis of applicable
content profile data for relevance to the query”; and “searching for information relevant to a
query associated with a first user.” In paragraph 63 of his report, Dr. Ungar states that in Culliss,
the “Internet articles are associated with key terms that they contain” citing column 3, lines 6064. Dr. Ungar goes on to state in paragraph 64 of his report that Culliss discloses that the
“articles are given a ‘key term score’ for each of the key terms that they contain.” Culliss 3:6566. These passages referenced by Dr. Ungar, however, refer solely to building the “initial index
setting.” Culliss 3:65-67. For all intents and purposes, Culliss’ rankings are based only on
popularity information.5
107.
Second, because Culliss does not disclose a content-based information
component, it does not disclose “combining pertaining feedback data . . . with the content profile
5
Culliss builds an index, consisting of a table (Culliss at 4:1-9) whose rows are items and whose
columns are terms (i.e., key words). Then when a user clicks on a displayed item shown in
response to a query, Culliss updates the entries in that table, for instance by adding +1 to each
cell corresponding the combination of a clicked item and query term (Culliss at 4:41-49). Culliss
also discloses a more elaborate updating mechanism (Culliss at 5:45-6:14) where he records both
the hits and displayed articles not clicked on for each term of each query. All updating
mechanisms are purely based on popularity of items by multiple users who issued queries
containing the terms in the table. Culliss initializes the table prior to initiating the updates based
on popularity. The update step is popularity based. Over time, the popularity counts totally
dominate the initialization (e.g., if an item was clicked 1000 times when a given terms was in the
query, the entry in the table would be 1000 + 1). In essence, the effect of the initialization
disappears over time -- it would be 1/10th or 1 percent in the above example -- and diminishing
further over time. Hence as the Culliss method is used it becomes for all intents and purposes a
popularity-only method.
25
data”; “combining the information from the feedback system with the information from the
scanning system”; or “combining the information found to be relevant to the query by other users
with the searched information.”
108.
Third, Culliss does not disclose “filtering each informon for relevance to the
query” or “filtering the combined information for relevance to at least one of the query and the
first user.” In paragraph 131 of his report, Dr. Ungar states that “Culliss ranks search results for
relevance to a query by calculating their aggregate key term scores for the terms in that query (id.
at 5:2-10), and each key term score is based on a combination of feedback data and content data
Culliss’ disclosure of ranking, however, does not disclose filtering.” Ranking is different than
filtering, which does not use a ranked list, but rather is an item-by-item process that considers
each item for exclusion based on its own attributes. Dr. Ungar’s example set forth in paragraphs
132-134 of his report does not cure or clarify the deficiencies of paragraph 131. Moreover, the
alternative described at column 13, lines 35-49 does not provide sufficient technical details to
explain how to combine the two systems and does not disclose filtering.
109.
Dr. Ungar in paragraph 140 states that “ranking a set of search results is sufficient
to meet the ‘filter’ limitation even if no candidate search results are excluded.” For the reasons
stated in the prior paragraph, I disagree. Filtering is an item-by-item process that considers each
item for exclusion.
110.
With respect to claims 14 and 15 of the ‘420 Patent, although Dr. Ungar states
that Culliss anticipates claims 14 and 15 of the ‘420 patent, I disagree. For the reasons already
noted in connection with claim 10, from which claims 14 and 15 depend, claims 14 and 15 are
not anticipated by Culliss.
26
Disclaimer: Justia Dockets & Filings provides public litigation records from the federal appellate and district courts. These filings and docket sheets should not be considered findings of fact or liability, nor do they necessarily reflect the view of Justia.
Why Is My Information Online?