Ex Parte Liang et alDownload PDFPatent Trial and Appeal BoardFeb 12, 201814231238 (P.T.A.B. Feb. 12, 2018) Copy Citation United States Patent and Trademark Office UNITED STATES DEPARTMENT OF COMMERCE United States Patent and Trademark Office Address: COMMISSIONER FOR PATENTS P.O.Box 1450 Alexandria, Virginia 22313-1450 www.uspto.gov APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO. 14/231,238 03/31/2014 Jisheng Liang FEVER-8822.US02 1016 22918 7590 02/14/2018 PERKINS COIE LLP - PAO General P.O. BOX 1247 SEATTLE, WA 98111 -1247 EXAMINER CHANNAVAJJALA, SRIRAMA T ART UNIT PAPER NUMBER 2157 NOTIFICATION DATE DELIVERY MODE 02/14/2018 ELECTRONIC Please find below and/or attached an Office communication concerning this application or proceeding. The time period for reply, if any, is set in the attached communication. Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the following e-mail address(es): patentprocurement @perkinscoie. com PTOL-90A (Rev. 04/07) UNITED STATES PATENT AND TRADEMARK OFFICE BEFORE THE PATENT TRIAL AND APPEAL BOARD Ex parte JISHENG LIANG, KRZYSZTOF KOPERSKI, JENNIFER COOPER, and THEODORE DIAMOND Appeal 2016-008126 Application 14/231,238 Technology Center 2100 Before ERIC S. FRAHM, NORMAN H. BEAMER, and STACY B. MARGOLIES, Administrative Patent Judges. FRAHM, Administrative Patent Judge. DECISION ON APPEAL STATEMENT OF THE CASE Introduction Appellants appeal under 35 U.S.C. § 134(a) from a Final Rejection of claims 1-20. Claim 9 was canceled in an after-final Amendment filed November 2, 2015. See Advisory Action mailed Nov. 12, 2015 (hereinafter, “Adv. Act.”), 1, box 7; Br. 1, 2. Therefore, the rejection of claims 1-8 and 10-20 is before us on appeal. We have jurisdiction under 35 U.S.C. § 6(b). We affirm the § 102(b) rejection of claims 1-8, 10, 14, 15, and 17-20, and reverse the § 102(b) rejection of claims 11-13 and 16. Appeal 2010-003268126 Application 14/231,238 Disclosed Invention and Exemplary Claim Appellants’ disclosed invention relates to a computer-implemented method and system for recommending content items (web pages, images, videos, articles, news stories, etc.) to users based on categories (sports, business, entertainment, actor, etc.) of the content items. Spec. 2, 4, 12. “A content item may reference an entity by naming or describing the entity. Entities may include people, places (e.g., locations), organizations (e.g., political parties, corporations, groups), events, concepts, products, substances, and the like.” Spec. ^ 13. A repository of entities and concepts is automatically built from various news feeds, databases, public or private repositories, etc., and the “entities may be organized into taxonomic hierarchies, based on taxonomic relations such as is-a, part-of, member-of, and the like.” Id. 17. A “taxonomic graph 430 is a tree that represents a hierarchy of categories” (Spec. ^ 48), and the taxonomic graph 430 shows links representing relationships of the categories. See Fig. 4D; Spec. 17, 48. In response to a user request for recommended content based on a category indicated by the user, “one or more content items that each have a corresponding category that matches the indicated category” are recommended. Abstract; see also Spec. 12-15. Independent claim 1, which is reproduced below with bracketed lettering and emphases added to denote the disputed limitation, is exemplary of the subject matter on appeal: 1. A computer-implemented method in a content recommendation system, the method comprising: processing a corpus of content items to determine, for each of the content items, multiple corresponding entities 2 Appeal 2010-003268126 Application 14/231,238 referenced by the content item, each of the determined entities being electronically represented by the content recommendation system; determining, for each of at least some of the content items, at least one corresponding category that is part of a taxonomy represented as a graph stored by the content recommendation system and that is associated with one of the multiple corresponding entities referenced by the content item, [A] wherein determining the at least one corresponding category includes aggregating common nodes in taxonomic paths that are associated respectively with a first determined entity and a second determined entity that are part of the graph such that the corresponding category relates the first and second determined entities in an is-a relation, a part- of relation, or a member-of relation; and storing, for each of the content items, the determined multiple corresponding entities and the determined at least one corresponding category. The Examiner’s Rejection The Examiner rejected claims 1-8 and 10-20 under 35 U.S.C. § 102(b) as being anticipated by Marchisio (US 7,398,201 B2; issued July 8, 2008). Final Act. 5-15; Adv. Act. 2; Ans. 2-7. Appellants ’ Contentions Appellants contend, inter alia, Marchisio fails to disclose limitation [A] recited in independent claim 1 (Br. 6-10). More specifically, Appellants contend Marchisio’s parse tree does not correspond to the taxonomic path recited in claim 1. Br. 7-10. Appellants also argue that Marchisio does not disclose aggregating common nodes in taxonomic paths as claimed. Br. 6- 7, 9-10. 3 Appeal 2010-003268126 Application 14/231,238 Appellants also argue dependent claims 6, 11-13, and 16 separately, contending that the various additional limitations recited in these claims are not disclosed by Marchisio (Br. 10-15). Issues on Appeal Based on Appellants’ arguments in the Appeal Brief (Br. 6-15), the following issues are presented on appeal1: (1) Did the Examiner err in rejecting claims 1-8, 10, 14, 15, and 17- 20 under 35 U.S.C. § 102(b) because Marchisio fails to disclose: (a) disputed limitation [A], including “aggregating common nodes in taxonomic paths,” as recited in representative independent claim 1; and/or (b) “ranking leaf node categories of taxonomic paths,” as recited in dependent claim 6? (2) Did the Examiner err in rejecting claims 11-13 and 16 under 35 U.S.C. § 102(b) because Marchisio fails to disclose the disputed limitations of claims 11-13 and 16, namely: (a) ranking content items based on a “credibility score” for each content item, as recited in claim 11; 1 Appellants present arguments as to independent claim 1, and rely on these arguments as to the patentability of independent claims 17 and 19 and dependent claims 2-5, 7, 8, 10, 14, 15, 18, and 20 (Br. 6-10, 15). Appellants present separate arguments for dependent claims 6, 11-13, and 16 (Br. 10-15). Based on Appellants’ arguments in the Appeal Brief and the similarity of subject matter among independent claims 1,17, and 19, we select claim 1 as representative of the group of claims consisting of claims 1-5, 7, 8, 10, 14, 15, and 17-20. 4 Appeal 2010-003268126 Application 14/231,238 (b) ranking content items based on “recency of each content item,” as recited in claim 12; (c) “collapsing similar content items into groups of content items,” as recited in claim 13; and (d) “determining popular entities for an indicated category,” as recited in claim 16? ANALYSIS We have reviewed the Examiner’s rejection in light of Appellants’ contention in the Appeal Brief (Br. 6-15) that the Examiner has erred, as well as the Examiner’s response to Appellants’ arguments (Ans. 2-7). We agree with Appellants’ arguments as to the anticipation rejection of claims 11-13 and 16 with regard to Marchisio (Br. 12-15) as applied by the Examiner (see Ans. 5-7). However, with regard to the Examiner’s anticipation rejection of representative claim 1 and dependent claim 6 with regard to Marchisio, we agree with the Examiner’s conclusions that Marchisio discloses (i) limitation [A], including aggregating common nodes in taxonomic paths, as recited in representative independent claim 1 (see, e.g., Final Act. 6-9; Ans. 3 4); and (ii) “ranking leaf node categories of taxonomic paths” as recited in claim 6 (Final Act. 10-11; Ans. 5). Anticipation Rejection of Claims 1—8, 10, 14, 15, and 17—20 Issue 1(a): Representative Claim 1 With regard to representative claim 1, we disagree with Appellants’ arguments. We adopt as our own (1) the findings and reasons set forth by the Examiner in the Final Office Action from which this appeal is taken 5 Appeal 2010-003268126 Application 14/231,238 (Final Act. 6-9), as well as the Advisory Action (Adv. Act. 2, i), and (2) the reasons set forth by the Examiner in the Examiner’s Answer (Ans. 3- 4) in response to Appellants’ Appeal Brief. We concur with the conclusions reached by the Examiner regarding the obviousness of the computer- implemented method for recommending content by a category for content items and entities as set forth in representative claim 1. Specifically, we agree with the Examiner that Marchisio discloses “aggregating common nodes in taxonomic paths” as recited in claim 1 by disclosing grouping nodes (see Final Act. 7; Advisory Act. 2, i (citing Marchisio, col. 14,11. 10-52)). Although Marchisio does not use the term “aggregate,” one of ordinary skill in the art would have understood Marchisio’s grouping of verbs based on the similarity of verb meanings/categories (see Marchisio, col. 14,11. 43^45) to be aggregating nodes as claimed. This is further amplified by the Examiner’s reasonable comparison (see Ans. 3—4; Final Act. 6-7, 9) of Marchisio’s parse tree shown in Figure 19 to Appellants’ taxonomic graph tree shown in Figure 4D (see Spec. 48-50 (describing Fig. 4D)). We agree with the Examiner’s findings (Final Act. 6-7) that paragraphs 13-15 (cited by the Examiner as pages 3^1 of the Specification), as well as paragraph 30, of Appellants’ Specification describe taxonomic relations of entities associated with categories as being linked in a semantic network, taxonomy, or graph. We also agree with the Examiner’s finding that Marchisio’s Figure 19 shows a hierarchy of categories (e.g., sentence, noun phrase, verb phrase, part of speech) of semantic relations including a top/root node 1901 and child/leaf nodes 1902-1911 on several levels below 6 Appeal 2010-003268126 Application 14/231,238 (see Marchisio, col. 20,11. 36-57) that corresponds to the recited “taxonomy represented as a graph” having “nodes” arranged in “taxonomic paths” as recited in claim 1. This is further supported by Appellants’ description of Figure 4D as a “taxonomic tree” illustrating a “taxonomic graph 430” that Appellants describe as “a tree that represents a hierarchy of categories that each have zero or more child categories connected via an arc or link representing a relation” beginning “with a unique root category” (Spec. ^ 48). Furthermore, one of ordinary skill in the art would have understood “taxonomy” to mean a “classification into ordered categories.” See http://ymw.dictionary.com/browse/taxonomy7s~t (noun, 2) (last visited Jan. 31, 2018). Therefore, the Examiner’s finding that Marchisio’s parse tree or “graphical representative of an example syntactic structure” (col. 5,11. 29-30 describing Figure 19; see also col. 20,11. 21-57) is “a hierarchical data structure (e.g., a tree) of syntactic units” (col. 2,11. 35-36), data objects, entity tags, etc. (see col. 2,11. 42-59) showing nodes that represent categories arranged in taxonomic paths (i.e., a hierarchy of categories) is reasonable. In this light, we do not agree with Appellants’ arguments that (i) “[t]he superficial similarities between Marchisio’s parse tree and a figure in Appellant’s Specification [Figure 4D] is not a sufficient basis to support an anticipation rejection” (Br. 9); and (ii) “[t]he fact that a reference uses the same word (e.g., ‘tree’ or ‘graph’) as a claim element under examination is not sufficient as a teaching or suggestion of that claim element” (Br. 10). To the contrary, the Examiner has made a prima facie case of anticipation of the disputed subject matter (limitation [A]) recited in claim 1, and Appellants 7 Appeal 2010-003268126 Application 14/231,238 have not adequately rebutted the Examiner’s findings. Based on the foregoing, as well as our discussion of Marchisio’s grouping versus Appellants’ recited aggregating, we are not persuaded by Appellants’ contentions (see Br. 6-10) that Marchisio “does not describe the concept of ‘aggregating common nodes in taxonomic paths’” as recited in claim 1 (Br. 6). Appellants’ contention (Br. 10) that the Examiner has modified or rearranged the elements of Marchisio’s parse tree to map them to the subject matter of Appellants’ claims in an impermissible manner is unpersuasive. Marchisio shows in Figure 19 a top node 1901 of taxonomic paths leading down through groups of leaf/category nodes arranged in a hierarchy of levels which corresponds to Appellants’ taxonomic graph 430 shown in Figure 4D, including root node 432 and leaf nodes below arranged in levels (e.g., levels 434a-c, 436a, etc.), which “is a tree that represents a hierarchy of categories” beginning with a unique root category” (Spec. ^ 48). Therefore, Appellants have not shown error, with either evidence or argument, in the Examiner’s anticipation rejection of claim 1. Accordingly, we sustain the Examiner’s anticipation rejection of representative claim 1, as well as claims 2-5, 7, 8, 10, 14, 15, and 17-20 grouped therewith. Issue 1(b): Claim 6 Dependent claim 6 further limits claim 1 by reciting: wherein determining the at least one corresponding category includes ranking leaf node categories of taxonomic paths associated with the determined multiple corresponding entities, the ranking based on the quantity of entities having a 8 Appeal 2010-003268126 Application 14/231,238 particular category and/or the a rank of an entity in a ranked list of entities. Claim 6 (emphasis added). Appellants argue that Marchisio does not disclose ranking leaf node categories of taxonomic paths (Br. 10-12). Appellants admit that “Marchisio discusses ordering search results” (Br. 11; see also Br. 12), but dispute the Examiner’s determination that ordering search results such as sentences discloses ranking leaf node categories as claimed. For the reasons that follow, we disagree with Appellants’ arguments that Marchisio fails to disclose the added limitations of claim 6. We adopt as our own (1) the findings and reasons set forth by the Examiner in the Final Office Action from which this appeal is taken (Final Act. 10-11), as well as the Advisory Action (Adv. Act. 2, d), and (2) the reasons set forth by the Examiner in the Examiner’s Answer (Ans. 5) in response to Appellants’ Appeal Brief. We add the following for emphasis. We agree with the Examiner (Ans. 5) that Marchisio (Figs. 19, 21; col. 20,11. 36-57; col. 40,11. 30-39) discloses ranking leaf node categories of taxonomic paths as claimed. Marchisio’s Figure 19 shows, for example, a top or root node 1901 having the category of “Sentence,” which is parsed into leaf nodes 1902 and 1903, child node 1902 having the category “Noun Phrase” and child node 1903 having the category “Verb Phrase.” The taxonomic path for leaf/child node 1903 continues to second level leaf/child node 1907 having the category “Noun Phrase,” and ends at node 1911 having the category “Noun” which is an entity. Marchisio, Fig. 19; col. 20, 9 Appeal 2010-003268126 Application 14/231,238 11. 36-57. Therefore, Marchisio discloses leaf node categories arranged in taxonomic paths. We also agree with the Examiner (Final Act. 10-11) that Marchisio’s disclosure of sorting query results based on designated document attribute values using weights to rank results (Marchisio, col. 39,11. 4-10; col. 40,11. 24-39) discloses ranking the results shown in Marchisio’s Figure 19 (i.e., leaf node categories), as claimed. Column 39, lines 4-10 and column 40, lines 24-39 of Marchisio (relied on by the Examiner in the Final Rejection) disclose ranking results or “resultant objects (e.g.., sentences)” (col. 39,1. 5). Marchisio describes parsing a query “using a variety of parsing levels and search [ing] the indexed stored data set to locate data [i.e., entities] that contains similar terms used in similar grammatical roles and/or with similar entity tag types [i.e., categories] as indicated by the query” (Abstract). Marchisio similarly describes returning “results to a query in an order that indicates responsiveness to the query” and that “the ordering of results is based upon weightings of the matched results from the data set,” which in turn can be based on “the degree of matching” (Marchisio, col. 4,11. 26-34) (emphases added)). Thus, Marchisio’s query results, which can be categories or entity tag types, are ranked. We also agree with the Examiner (Ans. 5) that Marchisio discloses that “the results are sorted based on document attribute values designated, for example, by a user” (i.e., a category) (col. 40,11. 30- 31). Further, Marchisio discloses users can designate subjects and/or objects for queries (col. 38,1. 55-col. 39,1. 1), and the subject and/objects include categories. And, Marchisio also discloses assigning weights to the resultant 10 Appeal 2010-003268126 Application 14/231,238 objects (col. 39,11. 4-8) to order/rank the results. In view of the foregoing, Appellants’ arguments (Br. 11) that Marchisio does not disclose ranking leaf node categories, and thus claim 6 is not met by Marchisio, are not persuasive. Accordingly, we sustain the Examiner’s anticipation rejection of dependent claim 6. Anticipation Rejection of Claims 11—13 and 16 Issue 2(a): Claim 11 Claim 11 further limits claims 1 and 8 by reciting: wherein selecting the one or more content items includes ranking the one or more content items based on a credibility score determined for each content item. Claim 11 (emphases added). Appellants’ Specification describes determining a credibility score/weight for content items (Spec. 65, 66; original claim 11) based on the credibility of the source of the content item. In the Final Rejection (see Final Act. 12), the Examiner finds Marchisio discloses ranking based on a credibility score for content items at column 35, line 56 through column 36, line 42. In the Advisory Action (Adv. Act. 2, e), the Examiner finds Marchisio’s Figures 39A and 39B additionally disclose the credibility score limitation recited in claim 11. And, in the Answer, the Examiner finds Marchisio’s column 35, line 58 through column 36, line 6, as well as column 36, lines 9^42 and 51-67, “strongly support[] generating and ordering parameter strings typically having terms including order/sort method” (Ans. 5). Notably, however, the Examiner addresses Marchisio’s disclosure of “polysemy” and 11 Appeal 2010-003268126 Application 14/231,238 ordering/sorting parameter strings (Ans. 5), but does not explain how/why the cited portions of Marchisio disclose ranking content items based on a credibility score as recited in claim 11. Appellants contend (Br. 12) Marchisio fails to disclose ranking content items based on a credibility score as claimed, and Marchisio instead describes a routine for sorting parameter strings used to construct queries in order of increasing ambiguity which is based on inverse document frequency or polysemy (i.e., proxies for ambiguity). Our review of Marchisio reveals Figures 39A and 39B (described in columns 37-40 of Marchisio) show a process for weighting verbs in sentences to determine a similarity weight to provide “contextually accurate results” (col. 35,1. 61), “in order of increasing ambiguity” (col. 35,1. 67), such that “the parameter strings comprising the terms with the least ambiguity of meaning (the lower polysemy values) are given a higher weight because they are most apt to return pertinent results to the . . . query” (col. 36,11. 39^12). We agree with Appellants that Marchisio discloses ordering search results (Br. 11-12), . And, while (i) Marchisio discloses determining an ambiguity score to determine whether results (i.e., categories and/or sentences) are accurate/relevant (see Marchisio col. 35,1. 56-col. 37,1. 1); and (ii) Appellants admit disambiguation or determining an ambiguity score is well known in the art (see Spec. ^ 27), Marchisio is silent as to determining a credibility score for content items based on the source of the content item as described by Appellants. The Examiner has not sufficiently explained how Marchisio’s ambiguity score for categories/sentences 12 Appeal 2010-003268126 Application 14/231,238 corresponds to the credibility score for content items (e.g., news items, articles, videos, websites, etc.) recited in claim 6. As a result, we are constrained by the record before us to find the Examiner erred in rejecting claim 11 as being anticipated by Marchisio. Issue 2(b): Claim 12 Claim 12 further limits claims 1 and 8 by reciting: wherein selecting the one or more content items includes ranking the one or more content items based on recency of each content item, such that more recent content items are ranked higher than less recent content items. Claim 12 (emphases added). Appellants’ Specification describes ranking article groups by various factors, such as “publication dates of the articles, preferring recently published articles” (Spec. ^ 72), and determining “recent entities for a given category” such as by scoring entities “based on the total number of recent articles (in the content store) that mention the entity and/or user visit traffic to the entity’s page” (Spec. ^ 73). Appellants’ Specification also states “[rjecency weight may be based on the difference between the publishing date of a document and the date when ingestion started” (Spec. ^ 66). The Examiner determines in the Final Rejection that Marchisio discloses ranking based on the recency of each content item at column 4, lines 13 through 23 (Final Act. 12; see also Adv. Act. 2, f (relying on same disclosure)). And, in the Answer the Examiner determines Marchisio’s column 36, lines 24 42 and column 40, lines 24-39 additionally disclose ranking results based on term frequency and/or polysemy (see Ans. 6). Notably, however, the Examiner (see Ans. 6) addresses Marchisio’s 13 Appeal 2010-003268126 Application 14/231,238 disclosure of “polysemy” and ordering/sorting parameter strings, but does not explain how/why the cited portions of Marchisio correspond to ranking content items based on recency as recited in claim 12. Appellants contend (Br. 13) Marchisio fails to disclose ranking content items based on recency as claimed, and Marchisio instead describes a routine for ordering/sorting results based on polysemy, term frequency, or weightings based on parts of speech. We agree. Marchisio is silent as to determining any recency of results. And, the Examiner has not sufficiently explained how Marchisio’s disclosure of searching for similar sentences (see col. 4,11. 13-23), determining relevancy based on polysemy or ambiguity score (see col. 36,11. 24—42), and/or sorting based on weights (see col. 40,11. 24-39) disclose ranking based on a recency of content items (e.g., news items, articles, videos, websites, etc.) as recited in claim 12. As a result, we are constrained by the record before us to find the Examiner erred in rejecting claim 12 as being anticipated by Marchisio. Issue 2(c): Claim 13 Claim 13 further limits claims 1 and 8 by reciting: wherein selecting the one or more content items includes collapsing similar content items into groups of content items, wherein similarity between two content items is based on at least one of: distance between signatures of the two content items, amount of overlap between titles of the two content items, amount of overlap between summaries of the two content items, amount of overlap between URLs referencing the two content items, and publishers of the two content items. Claim 13 (emphasis added). 14 Appeal 2010-003268126 Application 14/231,238 The Examiner determines in the Final Rejection that Marchisio discloses collapsing similar content items into groups of content items at column 12, lines 25 through 34 (Final Act. 12), and similarly determines in the Advisory Action (Adv. Act. 2, g) that Marchisio disclosed the collapsing limitation of claim 13 at column 12, lines 9 through 34, which describes a method of judging similarity between data set contents. And, in the Answer the Examiner determines Marchisio’s Figure 20A and column 22, line 64 through column 23, line 12 additionally disclose summarizing content items to indicate overlap of the content items (see Ans. 6). Notably, however, the Examiner addresses Marchisio’s disclosure of determining similarity of items, but does not explain how/why the cited portions of Marchisio disclose collapsing similar content items into groups of content items as recited in claim 13. Appellants contend (Br. 14) Marchisio fails to disclose collapsing similar content items into groups as claimed, and Marchisio instead describes a process for returning a list of documents that are similar to a selected paragraph. We agree. Marchisio is silent as to collapsing content items into groups, and simply discloses determining the similarity of data sets by finding (i) documents or content items in a data set that contain similar content to selected sentences of a user’s query (col. 12,11. 19-25); and (ii) an overlap of content (see, e.g., col. 12,11. 25-34). The Examiner has not sufficiently explained how Marchisio’s search for similar sentences, even based on similarity and/or overlap, is collapsing similar content items into groups based on overlap as recited in claim 13. In other words, although we agree with the Examiner that Marchisio discloses (i) 15 Appeal 2010-003268126 Application 14/231,238 determining a similarity of sentences based on overlap (see, e.g., col. 12,11. 25-34), and (ii) matching entities (i.e., content items) of a data set by entity tag type (i.e., category) (col. 23,11. 1-23), the Examiner has not adequately shown how Marchisio’s matching by similarities and/or overlap constitutes collapsing the matched sentences/categories into groups. And, even if Marchisio’s matching based on similarity were the recited collapsing into groups, we find Marchisio matches sentences or categories, and not “content items” as recited in claim 13 (see Spec. ^ 36 describing “content items” as a news item, report, image, audio source, video stream, etc.). As a result, we are constrained by the record before us to find the Examiner erred in rejecting claim 13 as being anticipated by Marchisio. Issue 2(d): Claim 16 Claim 16 further limits claim 1 by further reciting, inter alia: determinins popular entities for an indicated category, the popular entities having recently received an increased number of references by content items in the corpus and/or having more references by content items in the corpus than other entities. Claim 16 (emphases added). Appellants describe determining popular entities in a category by ranking entities from a specific category using a popularity score based on entity occurrences in top articles or number of occurrences for the entity in an entity store (Spec. ^ 73). The Examiner determines in the Final Rejection that Marchisio discloses determining popular entities for categories at column 9, line 54 through column 10, line 3 (Final Act. 13; see also Adv. Act. 2, h (relying on same disclosure)). And, in the Answer, the Examiner determines 16 Appeal 2010-003268126 Application 14/231,238 Marchisio’s column 10, lines 15 through 19 additionally discloses determining popular entities as claimed (see Ans. 6-7). Notably, however, the Examiner addresses Marchisio’s disclosure of determining the presence of keywords in grammatical structures (e.g., sentences), but does not explain how/why the cited portions of Marchisio disclose determining the popularity of entities as recited in claim 16. In other words, the Examiner has not sufficiently shown how/why Marchisio’s keyword matching of entities with entity tags, categories, or grammatical roles disclosed in columns 9-10 is the same as determining that the entity found is “popular” as recited in claim 16. Appellants contend (Br. 14-15) Marchisio fails to disclose determining popular entities as recited in claim 16. We agree for the reasons explained above. As a result, we are constrained by the record before us to find the Examiner erred in rejecting claim 16 as being anticipated by Marchisio. Summary In view of the foregoing, we do not sustain the Examiner’s anticipation rejection of claims 11-13 and 16 based on Marchisio. CONCLUSIONS (1) The Examiner did not err in rejecting claims 1-8, 10, 14, 15, and 17-20 as being anticipated under 35 U.S.C. § 102(b) by Marchisio because Marchisio discloses limitation [A] as recited in representative independent claim 1; and ranking leaf node categories as recited in dependent claim 6. (2) The Examiner erred in rejecting claims 11-13 and 16 as being anticipated under 35 U.S.C. § 102(b) by Marchisio because Marchisio fails 17 Appeal 2010-003268126 Application 14/231,238 to disclose the disputed limitations recited in dependent claims 11-13 and 16. DECISION The Examiner’s anticipation rejection of (i) claims 1-8, 10, 14, 15, and 17-20 is affirmed, and (ii) claims 11-13 and 16 is reversed. No time period for taking any subsequent action in connection with this appeal may be extended under 37 C.F.R. § 1.136(a)(l)(iv). AFFIRMED-IN-PART 18 Copy with citationCopy as parenthetical citation