Ex Parte Odom

Patent Trial and Appeal BoardMay 16, 2013

11194766 (P.T.A.B. May. 16, 2013)

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. 11/194,766 08/02/2005 Paul S. Odom 088245-5326 8638 23524 7590 05/17/2013 FOLEY & LARDNER LLP 3000 K STREET N.W. SUITE 600 WASHINGTON, DC 20007-5109 EXAMINER BROWN, ALVIN L ART UNIT PAPER NUMBER 3682 MAIL DATE DELIVERY MODE 05/17/2013 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 1 ___________ 2 3 BEFORE THE PATENT TRIAL AND APPEAL BOARD 4 ___________ 5 6 Ex parte PAUL S. ODOM 7 ___________ 8 9 Appeal 2011-004390 10 Application 11/194,766 11 Technology Center 3600 12 ___________ 13 14 15 Before MURRIEL E. CRAWFORD, ANTON W. FETTING, and 16 MEREDITH C. PETRAVICK, Administrative Patent Judges. 17 18 FETTING, Administrative Patent Judge. 19 20 21 DECISION ON APPEAL22 Appeal 2011-004390 Application 11/194,766 2 STATEMENT OF THE CASE1 1 Paul S. Odom (Appellant) seeks review under 35 U.S.C. § 134 of a 2 final rejection of claims 1-55, the only claims pending in the application on 3 appeal. We have jurisdiction over the appeal pursuant to 35 U.S.C. § 6(b). 4 The Appellant invented a search engine that provides relevant 5 advertisements associated with search results (Spec., para. 0003). 6 An understanding of the invention can be derived from a reading of 7 exemplary claim 1, which is reproduced below (bracketed matter and some 8 paragraphing added). 9 1. A method comprising: 10 [1] a first computer system conducting a search 11 using a search constraint 12 to generate search results, 13 wherein the search constraint is received over a network, 14 and 15 wherein the search results include a set of data items 16 contained within a set of information; 17 [2] the first computer system determining a topic 18 for each of the search results 19 to produce a set of topics; 20 [3] the first computer system determining a set of significant 21 topics 22 from the set of topics; 23 and 24 [4] transmitting advertisements 25 that are related to the set of significant topics 26 over the network 27 for display. 28 29 30 1 Our decision will make reference to the Appellant’s Appeal Brief (“App. Br.,” filed Aug. 11, 2010) and Reply Brief (“Reply Br.,” filed Dec. 23, 2010), and the Examiner’s Answer (“Ans.,” mailed Oct. 26, 2010). Appeal 2011-004390 Application 11/194,766 3 The Examiner relies upon the following prior art: 1 Swahn US 2004/0049541 Al Mar. 11, 2004 Anderson2 US 2004/0093327 Al May 13, 2004 Desikan US 2005/0131758 Al Jun. 16, 2005 Kraft US 2006/0026013 Al Feb. 2, 2006 Claims 1-3, 5-17, 20-27, and 30-55 stand rejected under 35 U.S.C. § 2 103(a) as unpatentable over Anderson and Swahn. 3 Claim 4 stands rejected under 35 U.S.C. § 103(a) as unpatentable over 4 Anderson, Swahn, and Kraft. 5 Claims 18, 19, 28, and 29 stand rejected under 35 U.S.C. § 103(a) as 6 unpatentable over Anderson, Swahn, and Desikan. 7 8 ISSUES 9 The issues of obviousness turn primarily on the scope of the 10 limitations “determining a topic for each of the search results to produce a 11 set of topics,” and “determining a set of significant topics from the set of 12 topics.” 13 14 FACTS PERTINENT TO THE ISSUES 15 The following enumerated Findings of Fact (FF) are believed to be 16 supported by a preponderance of the evidence. 17 Facts Related to the Prior Art 18 Anderson 19 2 Anderson incorporates U.S. Provisional Patent Application 60/413,536, entitled "METHODS AND APPARATUS FOR SERVING RELEVANT ADVERTISEMENTS", filed on September 24, 2002 and listing Jeffrey A. Dean, Georges R. Harik, and Paul Buchheit as inventors by reference. Appeal 2011-004390 Application 11/194,766 4 01. Anderson is directed to allowing advertisers to put targeted ads on 1 any page on the web by (i) obtaining content that includes 2 available spots for ads, (ii) determining ads relevant to content, 3 and/or (iii) combining content with ads determined to be relevant 4 to the content. Anderson, para. 0014. 5 02. Figure 1 is a high level diagram of an advertising environment. 6 Advertisers may directly, or indirectly, enter, maintain, and track 7 ad information in the system. The ads may include embedded 8 information, such as a link, meta information, and/or machine 9 executable instructions. Ad consumers may submit requests for 10 ads to, accept ads responsive to their request from, and provide 11 usage information to, the system. Other entities may provide 12 usage information (e.g., whether or not a conversion or click-13 through related to the ad occurred) to the system. This usage 14 information may include measured or observed user behavior 15 related to ads that have been served. Anderson, para. 0030. 16 03. One example of an ad consumer is a general content server that 17 receives requests for content and retrieves the requested content in 18 response to, or otherwise services, the request. The content server 19 may submit a request for ads that may include a number of ads 20 desired and content request information, that in turn may include 21 the content itself, a category corresponding to the content or the 22 content request (e.g., arts, business, computers, arts-movies, arts-23 music, etc.), part or all of the content request, content age, content 24 type (e.g., text, graphics, video, audio, mixed media, etc.), and 25 geolocation information. Anderson, para. 0031. 26 Appeal 2011-004390 Application 11/194,766 5 04. The content server may combine the requested content with one or 1 more of the advertisements, which is then forwarded towards the 2 end user that requested the content. The content server may 3 transmit information about the ads and how, when, and/or where 4 the ads are to be rendered (e.g., position, click-through or not, 5 impression time, impression date, size, conversion or not, etc.) 6 back to the system 120. Anderson, para. 0032. 7 05. Another example of an ad consumer is a search engine. A search 8 engine may receive queries for search results. In response, the 9 search engine may retrieve relevant search results (e.g., from an 10 index of Web pages). Such search results may include, for 11 example, lists of Web page titles, snippets of text extracted from 12 those Web pages, and hypertext links to those Web pages, and 13 may be grouped into a predetermined number of (e.g., ten) search 14 results. Anderson, para. 0033. 15 06. Content-relevant ad serving operations may include relevance 16 information extraction and/or generation operations. Relevance 17 information may be considered as a topic or cluster to which an ad 18 or document belongs. Probabilistic Hierarchical Inferential 19 Learner (phil) clusters are among those ways to develop such 20 topics. Anderson, para. 0128. 21 07. A document may be associated with one or more ads by mapping 22 a document identifier (e.g., a URL) to one or more ads. For 23 example, the document information may have been processed to 24 generate relevance information, such as a cluster (e.g., a phil 25 cluster), a topic, etc. The matching clusters may then be used as 26 Appeal 2011-004390 Application 11/194,766 6 query terms in a large OR query to an index that maps topics (e.g., 1 a phil cluster identifiers) to a set of matching ad groups. The 2 results of this query may then be used as first cut set of candidate 3 targeting criteria. The candidate ad groups may then be sent to the 4 relevance information extraction and/or generation operations 5 (e.g., a phil server) again to determine an actual information 6 retrieval (IR) score for each ad group summarizing how well the 7 criteria information plus the ad text itself matches the document 8 relevance information. Estimated or known performance 9 parameters (e.g., click-through rates, conversion rates, etc.) for the 10 ad group may be considered in helping determine the best scoring 11 ad group. Anderson, para. 0132. 12 08. Once a set of best ad groups have been selected, a final set of one 13 or more ads may be selected using a list of criteria from the best 14 ad group(s). The content-relevant an ad server can use this list to 15 request that an ad be sent back if K of the M criteria sent match a 16 single ad group. If so, the ad is provided to the requester. 17 Anderson, para. 0133. 18 09. Performance information (e.g., a history of selections or 19 conversions per URL or per domain) may be fed back in the 20 system, so that clusters or Web pages that tend to get better 21 performance for particular kinds of ads (e.g., ads belonging to a 22 particular cluster or topic) may be determined. This can be used 23 to re-rank content-relevant ads such that the ads served are 24 determined using some function of both content-relevance and 25 performance. A number of performance optimizations may be 26 Appeal 2011-004390 Application 11/194,766 7 used. For example, the mapping from URL to the set of ad groups 1 that are relevant may be cached to avoid re-computation for 2 frequently viewed pages. Anderson, para. 0134. 3 10. Anderson incorporates U.S. Provisional Application Serial No. 4 60/413,536, entitled "METHODS AND APPARATUS FOR 5 SERVING RELEVANT ADVERTISEMENTS", filed on 6 September 24, 2002 and listing Jeffrey A. Dean, Georges R. 7 Harik, and Paul Buchheit as inventors by reference. Anderson, 8 para. 0001. 9 11. U.S. Provisional Application Serial No. 60/413,536 at page 13 10 describes terms that appear frequently in the target document 11 being assigned a relatively higher weight than those that appear 12 less frequently. 13 12. U.S. Provisional Application Serial No. 60/413,536 at page 12 14 describes determining a subset of topics from a broader range of 15 topics for the purpose of ad selection. 16 Swahn 17 13. Swahn is directed to Information Retrieval and Display Systems 18 (IRDS) enhancements that enable search, display, and review of 19 Network based information to become a fast efficient process. 20 Swahn, para. 0029. 21 22 ANALYSIS 23 We are not persuaded by the Appellant’s argument that 24 [t]he independent claims contain distinct features, including 25 "determining a topic for each of the search results to produce a 26 set of topics" and "determining a set of significant topics from 27 Appeal 2011-004390 Application 11/194,766 8 the set of topics" which are not disclosed in Anderson, as relied 1 upon by the Examiner…. Anderson instead discloses 2 associating document content or document topics with relevant 3 ads or ad groups, and subsequently determining a set of best ads 4 from a list of relevant ads. Swahn fails to remedy Anderson's 5 deficiencies. 6 Appeal Br. 14. The claim does not narrow the attributes of topics or the 7 manner of determining them for each search result. Similarly, the claim 8 does not narrow the manner of determining a set of significant topics from 9 the set of topics, or any maximum degree of correspondence the sets may 10 exhibit between themselves, and indeed the two sets may be congruent. 11 As the Examiner found, such topics in the form of categories are 12 determined by the search input and accordingly the search output based on 13 that input. In addition, content-relevant ad serving operations may include 14 relevance information extraction and/or generation operations. Relevance 15 information may be considered as a topic or cluster to which an ad or 16 document belongs, and these topics may be developed by Probabilistic 17 Hierarchical Inferential Learner (phil) clusters. The ad groups represented 18 by these topics are again sent to a phil server to score their relevance. Thus, 19 the scoring serves to distinguish different sets of ads as to their relevance to 20 topics. A final set of one or more ads may be selected using a list of criteria 21 from the best ad group(s). Thus, whatever set of topics distinguishes the 22 higher scoring ads forms the selected group. Again, the claim does not 23 narrow how the determination is performed. To select an ad in Anderson is 24 to determine related topics. Further, as the Examiner found, U.S. 25 Provisional Application Serial No. 60/413,536 incorporated into Anderson 26 by reference described determining a subset of topics from a broader range 27 of topics for the purpose of ad selection. Ans. 14. 28 Appeal 2011-004390 Application 11/194,766 9 As to separately argued dependent claims 11, 36, and 41, reciting 1 hierarchically ranking topics based on the frequency of occurrence of the 2 topic and identifying topics having a frequency higher than a threshold, the 3 Examiner found that U.S. Provisional Application Serial No. 60/413,536 4 described using such a frequency as a scoring mechanism, evidencing that it 5 was predictable for one of ordinary skill to use frequency of occurrence in 6 Anderson’s scoring mechanism. Ans. 16. The use of a threshold is implied 7 in any scoring mechanism, as otherwise there is no need to score. Appellant 8 did not respond to this finding in the Reply Brief. 9 10 CONCLUSIONS OF LAW 11 The rejection of claims 1-3, 5-17, 20-27, and 30-55 under 35 U.S.C. § 12 103(a) as unpatentable over Anderson and Swahn is proper. 13 The rejection of claim 4 under 35 U.S.C. § 103(a) as unpatentable 14 over Anderson, Swahn, and Kraft is proper. 15 The rejection of claims 18, 19, 28, and 29 under 35 U.S.C. § 103(a) as 16 unpatentable over Anderson, Swahn, and Desikan is proper. 17 18 DECISION 19 The rejections of claims 1-55 is affirmed. 20 No time period for taking any subsequent action in connection with 21 this appeal may be extended under 37 C.F.R. § 1.136(a). See 37 C.F.R. 22 § 1.136(a)(1)(iv) (2011). 23 24 AFFIRMED 25 26 27 Appeal 2011-004390 Application 11/194,766 10 hh 1