Ex Parte Gross

Patent Trial and Appeal BoardAug 5, 2013

11117783 (P.T.A.B. Aug. 5, 2013)

UNITED STATES PATENT AND TRADEMARKOFFICE 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. 11/117,783 04/28/2005 John N. Gross 088245-3328 1058 23524 7590 08/06/2013 FOLEY & LARDNER LLP 3000 K STREET N.W. SUITE 600 WASHINGTON, DC 20007-5109 EXAMINER KUDDUS, DANIEL A ART UNIT PAPER NUMBER 2164 MAIL DATE DELIVERY MODE 08/06/2013 PAPER 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. PTOL-90A (Rev. 04/07) UNITED STATES PATENT AND TRADEMARK OFFICE ____________ BEFORE THE PATENT TRIAL AND APPEAL BOARD ____________ Ex parte JOHN N. GROSS ____________ Appeal 2011-001067 Application 11/117,783 Technology Center 2100 ____________ Before CAROLYN D. THOMAS, DAVID M. KOHUT, and DAVID C. McKONE, Administrative Patent Judges. McKONE, Administrative Patent Judge. DECISION ON APPEAL Appellant appeals under 35 U.S.C. § 134(a) from the Final Rejection of claims 1-13 and 32-34. See App. Br. 2. Claims 14-31 are withdrawn. See id. We have jurisdiction under 35 U.S.C. § 6(b). We affirm-in-part. Appeal 2011-001067 Application 11/117,783 2 THE INVENTION Appellant’s invention relates to identifying and measuring changes in content, linking, and clustering of Internet web pages. See Spec. 1:10-13. Claim 1, which is illustrative of the invention, reads as follows: 1. A method of analyzing cross-linking between a set of web pages comprising: identifying a plurality of webpages; determining, with a computer device, a first cross- linking value between said plurality of webpages; comparing the determined first cross-linking value with a second cross-linking value using a cross-entropy analysis; and based on the comparison, determining whether one or more clusters of cross-linking exist between said plurality of webpages. THE REJECTIONS The Examiner relies on the following prior art in rejecting the claims: Ryan US 6,421,675 B1 July 16, 2002 Lei US 2004/0012625 A1 Jan. 22, 2004 Garber Tal Garber et al., From Density to Destiny: Using Spatial Analysis for Early Prediction of New Product Success (Feb. 2002) Claims 1, 4-7, 9, 11, and 32 stand rejected under 35 U.S.C. § 103(a) as unpatentable over Lei and Garber. See Ans. 4-6. Claims 2, 3, 8, 10, 12, 13, 33, and 34 stand rejected under 35 U.S.C. § 103(a) as unpatentable over Lei, Garber, and Ryan. See Ans. 6-8. Appeal 2011-001067 Application 11/117,783 3 ANALYSIS REJECTION OF CLAIMS 1, 4-7, 9, 11, and 32 UNDER 35 U.S.C. § 103(a) Claims 1, 4, 6, 11, and 32 The Examiner finds that Lei teaches “determining whether one or more clusters of cross-linking exist between said plurality of webpages,” as recited in claim 1. See Ans. 5. Appellant contends that this finding is erroneous. See App. Br. 7-9. Lei teaches a technique for identifying web pages to collect, or “hoard,” for later offline browsing. See Lei, Abstract. A user can specify a set of “base” uniform resource locators (“URLs”) and assign to them “interestingness” values between 0 and 1. Lei, ¶ 0023. Lei’s system hoards web pages corresponding to the base URLs as well as certain web pages reachable from (i.e., linked to) the base web pages. See Lei, ¶ 0024. A hyperlink to a “target” web page that is a candidate for hoarding can be represented by a “content feature vector,” which can be computed based on the text on the base web page immediately surrounding the hyperlink anchor element containing the URL for the target web page. See Lei, ¶ 0026. Lei’s system also creates an “attribute feature vector” for the hyperlink, with details such as the location, type, and size of the target web page. Lei, ¶ 0027. In deciding whether to hoard additional web pages beyond a base web page, Lei’s system iterates through all the hyperlinks embedded on the base web page, computes content feature vectors and attribute feature vectors for those hyperlinks, and computes interestingness values for those hyperlinks. See Lei, ¶ 0031. To compute an interestingness value for a target hyperlink on a base web page, Lei’s system compares the content feature vector and attribute feature vector of the target hyperlink with Appeal 2011-001067 Application 11/117,783 4 corresponding vectors for the base web page and corresponding vectors for an “interests profile” provided by the user. Lei, ¶¶ 0028, 0032. Interestingness values show the similarity between a target web page and both the base web page and the interests profile. See Lei, ¶¶ 0016, 0032. The Examiner finds that such computing of interestingness values and hoarding of groups of linked web pages based on the interestingness values is determining whether one or more clusters of cross-linking exist between the base web page and the linked web pages. See Ans. 9-10. The Examiner also finds that the base web page and linked web pages constitute a hierarchical cluster. See Ans. 10. Appellant contends that Lei’s “modeling user interests in webpages and the computation of an ‘interestingness value’” is distinguishable from “determining the existence of clusters of cross-linking between multiple webpages,” drawing a distinction between “modeling” and “determining.” Ans. 8-9. Appellant, however, does not adequately explain this distinction. Lei teaches retrieving two or more web pages related by hyperlinks based on a predicted interestingness determined, in part, from the hyperlinks. See Lei, ¶¶ 0031-32; Ans. 9-10. Appellant does not explain how Lei’s system can hoard a group of linked documents (an example of a “cluster”) without determining that such a cluster exists. Appellant further argues that a determination of an interestingness value “based on a comparison of content of a selected webpage to content of a home page and/or to a user interests profile” is not a determination of clusters of cross-linking between multiple web sites. App. Br. 9. However, Lei also teaches determination of an interestingness value based on the hyperlink to the selected webpage in the home page. See Lei, ¶¶ 0026, 0027, Appeal 2011-001067 Application 11/117,783 5 0032; Ans. 9-10. Appellant does not persuasively explain how this is different from determining the existence of a cluster of cross-linking. Appellant also argues that “‘Cross-linking’ refers to two or more webpages linking to each other, which is more than ‘linking’ which refers simply to a single link from one page to another.” App. Br. 9. However, Appellant does not explain the basis for this proposed construction. The Specification describes cross-linking as including links among web pages within a web site as well as links to and from a web page from external web sites. See Spec. 50:4-5; 50:31-51:3. Appellant does not point to anything in the Specification or otherwise limiting cross-linking to web pages linking to each other. Thus, Appellant has not persuasively shown that a base web page hyperlinking to additional web pages on the same website (see Lei, ¶¶ 0009, 0028, 0031) is distinguishable from the claimed cross-linking. In the Reply Brief, Appellant argues that a person of ordinary skill in the art would not have had a reasonable expectation of success combining Lei and Garber. See Reply Br. 7-9. Appellant could have made this argument in the Appeal Brief, but did not. Instead, the argument was first presented in the Reply Brief. Thus, the argument is waived. See, e.g., Ex parte Borden, 93 USPQ2d 1473, 1474 (BPAI 2010) (informative) (“[T]he reply brief [is not] an opportunity to make arguments that could have been made in the principal brief on appeal to rebut the Examiner’s rejections, but were not.”). In sum, Appellant has not persuasively shown that the Examiner erred in concluding that claim 1 would have been obvious. Accordingly, we sustain the rejection of: (1) independent claim 1; (2) independent claim 32, which contains commensurate limitations and is not argued separately with Appeal 2011-001067 Application 11/117,783 6 particularity; and (3) claims 4, 6, and 11, which depend on claim 1 and are not argued separately with particularity. Claim 5 The Examiner finds that Lei teaches “determining an age of at least one of said plurality of web pages.” See Ans. 6 (citing Lei, ¶¶ 0006, 0007, 0024, 0031). Appellant contends that Lei does not teach determining the age of a web page. See App. Br. 10. We agree with Appellant. None of the Examiner’s citations evidence determining the age of a web page. For example, paragraphs 0006 and 0024 address limiting the time consumed by hoarding, but do not address the amount of time a web page has existed. The other citations are similarly inapposite. Accordingly, we do not sustain the rejection of claim 5. Claim 7 Lei teaches updating a user’s interest profile by analyzing a demanded new web page and summing the content feature vector of the new web page with the content feature vector of the interests profile to generate a new content feature vector in the interests profile. See Lei, ¶ 0029. The interests profile is used to determine whether to hoard other web pages. See Lei, ¶ 0031-32. Similarly, Lei’s claim 7 recites “monitoring a user’s Web page browsing behavior, and dynamically updating said content and attribute vectors [of a user interests profile] based on the Web pages browsed by that user.” The Examiner finds that these teachings show “determining that at least one of said plurality of web pages is a trendsetter web page. See Ans. 6, 13-14. Appeal 2011-001067 Application 11/117,783 7 Appellant argues that “disclosing a way to determine if a user is interested in a website is not the same as ‘determining that at least one of said plurality of web pages is a trendsetter web page,’” and that “[a]ggregating content of Web documents is not the same as determining whether one page is a trendsetter web page.” App. Br. 11. These arguments, however, do not address the Examiner’s finding (Ans. 6, 13-14) that updating a user’s interests profile with based on an identified web page (Lei, ¶ 0029) teaches determining that the identified web page is a trendsetter web page. Thus, Appellant’s arguments are unpersuasive. Accordingly, we sustain the rejection of claim 7. Claim 9 Claim 9 depends on claim 1 and adds “determining that at least one of said plurality of webpages is a trend laggard web page.” Appellant contends that, according to the Specification, “trend laggard indicates the end of a useful adoption cycle and an inability to achieve a clustering effect.” App. Br. 12 (citing ¶ 0154 of the published Specification1). The Examiner concludes that the Specification treats a trend laggard as simply the opposite of a trend setter. See Ans. 14-15. For example, the Specification states that “if trend laggards (and/or trend rejecters) can be identified, their contributions or weightings to a recommender system might be adjusted in a similar manner to that provided for trendsetters, except in the opposite manner.” Spec. 30:21-23. According to the Examiner, Lei 1 Paragraph 0154 of the published Specification discusses trendsetters rather than trend laggards. The evidence suggests that Appellant intended to cite paragraph 0157, which corresponds to page 29, line 31-page 30, line 8, of the original Specification. Appeal 2011-001067 Application 11/117,783 8 recognized that prior solutions increased false predictions, which wasted resources. See Ans. 14 (citing Lei, ¶ 0007). The Examiner then points to paragraph 0023 of Lei, which teaches a user assigning interestingness values to specified base web pages, wherein a weighting of 0 indicates a web page that is the least interesting. See Ans. 15. Thus, Lei teaches assigning low interestingness values to web pages that are poor predictors of other web pages that the user will find interesting. Appellant does not explain why this is distinguishable from determining that a web page is a trend laggard web page. For example, assigning a low weight to a base web page would make it less likely that web pages it links to would be identified as interesting. Thus, even under Appellant’s construction of “trend laggard,” a low-weighted base web page would indicate that the web pages it links to should not be included in a cluster of cross-linked web pages. Accordingly, we sustain the rejection of claim 9. REJECTION OF CLAIMS 2, 3, 8, 10, 12, 13, 33, and 34 UNDER 35 U.S.C. § 103(a) Claims 2, 3, 8, 10, 33, and 34 Appellant argues that claims 2, 3, 8, 10, 33, and 34 are patentable for essentially the same reasons as claim 1. See App. Br. 13. For the reasons given above, we are not persuaded the Examiner erred in rejecting claim 1. Accordingly, we sustain the rejection of claims 2, 3, 8, 10, 33, and 34. Claims 12 and 13 Appellant contends that Ryan fails to teach an online auction website, or using a cross-linking value to identify an item of interest for marketing on Appeal 2011-001067 Application 11/117,783 9 an online auction website, as recited in claim 12. See Ans. 14. The Examiner cites to, inter alia, column 29, lines 10-18, of Ryan, finding that it discloses book sales. See Ans. 8. However, this passage teaches the identification of “book sales” as a suggested key-word for a search based on analysis of the behavior of other uses. See Ryan, col. 29, ll. 10-22. Likewise, the Examiner’s citation to column 34, lines 28-48, does not show marketing an item of interest on a website. Rather, it teaches adjusting the rankings of search results for keywords based on the amounts content providers are willing to pay. Thus, we agree with Appellant that Ryan does not teach “wherein said first cross-linking value is used to identify an item of interest for marketing on said online auction site.” Similarly, for claim 13, Appellant argues that Ryan fails to teach “wherein said first cross-linking value is used to identify individuals to be excluded from said online auction site to reduce bias in search queries.” See App. Br. 14. We agree with Appellant. While portions of Ryan, such as column 24, line 61-column 25, line 25, teach tailoring search results to the profile of a particular user (e.g., returning web sites related to rugby for a New Zealander’s search for “field sports” and returning web sites related to baseball for an American’s similar search), the Examiner has not shown that Ryan teaches using a cross-linking value to identify individuals to be excluded from an online auction. Accordingly, we do not sustain the rejection of claims 12 and 13. Appeal 2011-001067 Application 11/117,783 10 ORDER The decision of the Examiner to reject claims 1-4, 6-11, and 32-34 is affirmed. The decision of the Examiner to reject claims 5, 12, and 13 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)(1). See 37 C.F.R. § 1.136(a)(1)(iv). AFFIRMED-IN-PART gvw