Ex Parte GrossDownload PDFPatent Trial and Appeal BoardFeb 24, 201712858150 (P.T.A.B. Feb. 24, 2017) 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. 12/858,150 08/17/2010 JOHN N. GROSS JNG.2004 10D 5292 23694 7590 02/28/2017 Law Office of J. Nicholas Gross, Prof. Corp. PO BOX 9489 BERKELEY, CA 94709 EXAMINER STERRETT, JONATHAN G ART UNIT PAPER NUMBER 3623 NOTIFICATION DATE DELIVERY MODE 02/28/2017 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): eofficeaction @ appcoll.com jngross@pacbell.net anthonygreek @ gmail. com PTOL-90A (Rev. 04/07) UNITED STATES PATENT AND TRADEMARK OFFICE BEFORE THE PATENT TRIAL AND APPEAL BOARD Ex parte JOHN N. GROSS Appeal 2014-009012 Application 12/858,150 Technology Center 3600 Before MURRIEL E. CRAWFORD, JAMES A. WORTH, and ROBERT J. SILVERMAN, Administrative Patent Judges. CRAWFORD, Administrative Patent Judge. DECISION ON APPEAL STATEMENT OF THE CASE Appellant seeks our review under 35 U.S.C. § 134 of the Examiner’s final decision rejecting claims 1, 2, 4—13, and 18—21.1 We have jurisdiction over the appeal under 35 U.S.C. § 6(b). We REVERSE. Claim 1 is illustrative: 1. A method using a computing system of evaluating advertising associated with a recommender system incorporated as part of a search engine at a first website, 1 Claims 3 and 14—17 are cancelled. See Appeal Br., Claims App. Appeal 2014-009012 Application 12/858,150 which recommender system is used for recommending items of interest, the method comprising the steps of: (a) specifying with the computing system a first reference item (A) to be used in evaluating the recommender system; (b) identifying with the computing system a list of reference items presented by the recommender system to a first user as a plurality of separate recommendations; (c) automatically comparing said first reference item (A) with said list of items with the computing system to identify a correlation value exhibited by the recommender system between said first reference item (A) and at least one second item (B); (d) automatically adjusting advertising by a first vendor presented for said first reference item (A) at the first website with the computing system based on said correlation value with said second item (B); wherein said first vendor can exploit proxy advertising made by the recommender system for first reference item (A) as a result of such item being correlated and presented with second item (B) in said list of reference items at the first website. Appellant appeals the following rejection: 1. Claims 1, 2, 4—13, and 18—21 under 35 U.S.C. § 103(a) as unpatentable over Burke (Robin Burke, Hybrid Recommender Systems: Surveys and Experiments, 12 User Modeling and User-Adapted Interaction 331, 331— 370 (2002)) in view of Karande (Kiran W. Karande, The Effect of Brand Characteristics and Retailer Policies on Response to Retail Promotions: Implications for Retailers, 73(3) Journal of Retailing 249, 249—278 (1995)). 2 Appeal 2014-009012 Application 12/858,150 ISSUE Did the Examiner err in rejecting the claims because the prior art does not disclose automatically comparing a first reference item (A) with a list of items presented by the recommender system to identity a correlation value between the first item (A) and a second item (B)? ANALYSIS The Appellants argue that Burke does not identify a correlation value between a first item and a second item on a list presented by the recommender system. We agree. The Examiner, in the Final Office Action (“Final Act.”), relies on the second and third paragraphs, on page 357 of Burke, for teaching how the user’s ratings of restaurants are used to understand the similarity between the users and how the restaurants are similarly or differently scored (Final Act. 4). We find that Burke discloses a collaborative recommender system, which uses aggregate ratings or recommendations of objects, recognizes commonalities between users on the basis of their ratings, and generates new recommendations based on inter-user comparisons (Burke 332). The profiles of two users are converted to numeric vectors and then correlated, and a heuristic technique is used to bring out the similarity between two users. (Burke 357—358). There is no disclosure in this portion of Burke of identifying a correlation value between a first item and a second item that are on a list presented by the recommender system. In the Answer, the Examiner relies on other portions of Burke for teaching a correlation of items recommended to a single user. For example, the Examiner relies on Table 1 for teaching various recommendation 3 Appeal 2014-009012 Application 12/858,150 techniques that rely on the comparison of items. The Examiner specifically directs our attention to the teaching in Burke related to a utility-based recommendation that ranks items according to a user’s preferences and a knowledge-based recommendation that compares items vis-a-vis a user’s needs as clearly disclosing correlations between items in regard to how the items meet the user’s needs. The utility-based technique uses the utility function of items and applies that function to other items to determine their rank. The knowledge-based technique uses needs and interests of a user to infer a match between an item and the user’s needs. However, the claims require that a correlation value is computed between a first reference item on a list presented by a recommender system and a second item. The Examiner has not directed our attention to a teaching in Burke that the items compared in the utility-based recommendation and the knowledge-based recommendation are presented by a recommender system. In view of the foregoing, we will not sustain the Examiner’s rejection of claim 1 and claims 2, and 4—13 dependent therefrom. We will also not sustain the rejection as it is directed to claim 21 because this claim requires recommendations given in lists by the recommender system. We will also not sustain the rejection of claim 18 and claim 19 dependent therefrom because we agree with the Appellantthat the prior art does not disclose providing a second recommender system and identifying the frequency of a first set of recommendations given by the second recommender system. In this regard, we do not agree with the Examiner that claim 18 does not require a second recommender system because the claim includes “and/or” language. Claim 18 does indeed positively recite a second 4 Appeal 2014-009012 Application 12/858,150 recommender at a second website in line 11 of the claim. The and/or language in the claims relates to where the modification of the advertising can take place, i.e., at the first website or at the second website. In view of the foregoing, we will not sustain the rejection as it is directed to claim 18 and claims 19 and 20 dependent therefrom. DECISION The decision of the Examiner is reversed. REVERSED 5 Copy with citationCopy as parenthetical citation