Ex Parte Chidlovskii et alDownload PDFPatent Trial and Appeal BoardFeb 7, 201311156776 (P.T.A.B. Feb. 7, 2013) 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. 11/156,776 06/20/2005 Boris Chidlovskii 20041056USNP-XER00944US01 7754 27885 7590 02/08/2013 FAY SHARPE LLP 1228 Euclid Avenue, 5th Floor The Halle Building Cleveland, OH 44115 EXAMINER WRIGHT, ELIZABETH G ART UNIT PAPER NUMBER 2172 MAIL DATE DELIVERY MODE 02/08/2013 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 BORIS CHIDLOVSKII and JEROME FUSELIER ____________ Appeal 2010-007140 Application 11/156,776 Technology Center 2100 ____________ Before JOHN A. JEFFERY, BRUCE R. WINSOR, and JAMES B. ARPIN, Administrative Patent Judges. JEFFERY, Administrative Patent Judge. DECISION ON APPEAL Appellants appeal under 35 U.S.C. § 134(a) from the Examiner’s rejection of claims 1-20. We have jurisdiction under 35 U.S.C. § 6(b). We affirm. STATEMENT OF THE CASE Appellants’ invention classifies semi-structured documents by distinguishing sub-tree structural information as a distinct representative Appeal 2010-007140 Application 11/156,776 2 characteristic of a document fragment. See generally Abstract. Claim 1 is illustrative: 1. A method of classifying a semi-structured document, comprising: identifying the document to include a plurality of document fragments, wherein at least a portion of the fragments include a recognizable structure corresponding to fragment content comprising a sub-tree node tag, a sub-tree outer structure and a subtree inner structure; recognizing selected ones of the fragments to comprise pre- determined content and structure with reference to the node tag, the inner structure and the outer structure; and classifying the document as a particular type of document in accordance with the recognizing. THE REJECTIONS 1. The Examiner rejected claims 1-8 and 11-20 under 35 U.S.C. § 102(b) as anticipated by Lindblad (US 2004/0103091 A1; May 27, 2004). Ans. 4-10.1 2. The Examiner rejected claims 9 and 10 under 35 U.S.C. § 103(a) as unpatentable over Lindblad and Zha (US 7,028,027 B1; Apr. 11, 2006; filed Sept. 30, 2002). Ans. 11-12. THE ANTICIPATION REJECTION The Examiner finds that Lindblad discloses every recited element of representative claim 1 including recognizing selected document fragments to comprise predetermined content and structure with reference to a sub-tree’s node tag, inner structure, and outer structure, the latter of which is said to 1 Throughout this opinion, we refer to (1) the Appeal Brief filed June 8, 2009 (supplemented November 4, 2009) (“App. Br.”); (2) the Examiner’s Answer mailed February 19, 2010 (“Ans.”); and (3) the Reply Brief filed April 19, 2010 (“Reply Br.”). Appeal 2010-007140 Application 11/156,776 3 correspond to portions of Lindblad’s tree structure outside of a particular sub-tree in Figure 9. Ans. 4-5, 13-15. According to the Examiner, Lindblad classifies a document as a particular type in terms of its similarity according to this recognition. Ans. 5, 14-15. Appellants argue that not only are sub-tree outer structures irrelevant to Lindblad which discloses a generic tree with nodes and inner tree structures, there is no classification in Lindblad. App. Br. 12-16; Reply Br. 4-6. Appellants also argue various other limitations summarized below. ISSUES Under § 102, has the Examiner erred by finding that Lindblad discloses: (1)(a) recognizing selected document fragments to comprise predetermined content and structure with reference to a sub-tree’s node tag, inner structure, and outer structure, and (b) classifying the document as a particular type according to this recognition, as recited in claim 1? (2) recognizing semantic content of the fragment within the document as predetermined content, as recited in claim 2? (3) concatenating the fragment’s content components, as recited in claim 3? (4) recognizing a structural element of the fragment as the predetermined structure, as recited in claim 4? (5) identifying the fragment’s relative position within the document, as recited in claim 5? (6) identifying a logical structure of the fragment, as recited in claim 6? Appeal 2010-007140 Application 11/156,776 4 (7)(a) representing the fragment as a sub-tree having a navigational path between a fragment root and leaf, and (b) defining the logical structure as the navigational path, as recited in claim 7? (8) determining distinct likelihoods of correspondence to the type of document for the sub-tree outer and inner structures, as recited in claim 12? (9) combining the distinct likelihoods for estimating a final document type (claim 13), where the likelihoods are weighted by a pre-selected weight (claim 14)? (10) determining distinct likelihoods of correspondence to the type of document for each of the sub-tree outer structure, sub-tree inner structure, sub-tree content, and sub-tree node tags and attributes, as recited in claim 16? (11) combining the distinct likelihood [sic] for estimating a final document type, as recited in claim 17? ANALYSIS Claims 1, 8, 11, 15, and 18 We begin by noting that it is undisputed that Lindblad accesses both inner and outer tree structures, and that certain tags (e.g., “” in Lindblad’s Figure 9) distinguish the associated nodes as a separate inner tree section that is different from other nodes in other outer tree portions not within the particular distinguished section. Ans. 15; Reply Br. 5-6 (citing Lindblad ¶ 0049; Fig. 9). Nor is it disputed that Lindblad’s inner and outer structures distinguish a portion of the tree (i.e., the inner structure) that should be used to classify a document from the other irrelevant portion of Appeal 2010-007140 Application 11/156,776 5 the tree (i.e., the outer structure) that is irrelevant to the classification process. Ans. 15; Reply Br. 6 (citing Lindblad ¶ 0066; Fig. 14). The question, then, is whether the Examiner erred by finding that Lindblad (1) recognizes fragment content and structure with respect to this “outer structure” along with the sub-tree node tag and inner structure, and (2) classifies documents according to this recognition as claimed. On this record, we find no error in the Examiner’s position. First, the term “outer structure” is not defined in Appellants’ Specification, but rather is described somewhat generally with respect to an inner node sub-tree defined by a sub- tree root node 30 with a sub-tree inner structure 32, sub-tree content 34, and a sub-tree outer structure 36. Spec. ¶ 0031; Fig. 1(c). As shown in Appellants’ Figure 1(c) reproduced below, sub-tree outer structure 36 is a generally triangular region outside the speckled regions 30, 32 denoting the sub-tree root node and content. Appellants’ Figure 1(c) showing sub-tree outer structure 36 Appeal 2010-007140 Application 11/156,776 6 This outer structure is said to not include an outer path of the sub-tree structure, but rather pertains to its contextual relationship in the document. App. Br. 12 (citing Figs. 1 and 2(c) of the present application). Based on this description, we see no error in the Examiner’s reliance on the tree structure outside of a particular sub-tree delineated by dashed borders 52 in Lindblad’s Figure 9, for this recognized “outer” structure provides a contextual relationship for the sub-tree, at least with respect to its relationship to other nodes and sub-trees within hierarchical tree structure 50 that represents a multi-level XML document. See Lindblad ¶¶ 0048-49; Figs. 8-9. Lindblad’s Figure 9 showing sub-tree borders 52 in tree 50 is reproduced below: Lindblad’s Figure 9 showing sub-tree borders 52 in tree 50 Based on this inner and outer structural relationship with respect to particular sub-trees and their associated tree, we see no error in the Appeal 2010-007140 Application 11/156,776 7 Examiner’s position that Lindblad fully meets the recited recognition step. Appellants’ arguments regarding the particular outer structure in their Figure 2(c), including a child-to-father node vector and attributes associated with the structure (App. Br. 14-15; Reply Br. 4), are unavailing and not commensurate with the scope of the claim. Nor do we find error in the Examiner’s position that Lindblad classifies documents as particular types of documents—namely those that are more similar than others—according to Lindblad’s fragment recognition noted above. Ans. 5, 14 (citing Lindblad ¶ 0066; Fig. 14). Although Appellants assert that this document similarity determination does not comport with their “definition” of “classification” in the Specification, namely “distinguishing between three possible groups of features that will allow a representation of the document fragment comprising the sub-tree by its semantic and contextual content” (Reply Br. 5 (citing Spec. ¶ 0032)), we note that this passage merely describes classification in the context of Appellants’ Figure 2, but does not define the term to so limit its interpretation. That Appellants’ Specification clearly and unambiguously defines other terms in the Specification only bolsters this conclusion.2 We are therefore not persuaded that the Examiner erred in rejecting independent claim 1, and claim 8 not separately argued with particularity. We likewise sustain the Examiner’s rejection of independent claims 11 and 2 Compare Spec. ¶ 0018 (“A sub-tree is defined as a document fragment, rooted at some node in the document structure hierarchy.” (emphasis added)); id. (“By ‘path’ is meant the navigation from a root of the document to a leaf, i.e., the structure between the root and the leaf.” (emphasis added)); ¶ 002 (“By ‘semi-structured documents’ is meant a free-form (unstructured) formatted text which has been entered with meta information.” (emphasis added)). Appeal 2010-007140 Application 11/156,776 8 18 (Ans. 5-6) (and claim 15 not separately argued with particularity), for Appellants’ arguments (App. Br. 17-19) do not persuasively rebut the Examiner’s position for similar reasons. Claim 2 We also sustain the Examiner’s rejection of claim 2 reciting recognizing semantic content of the fragment within the document as the predetermined content. Appellants’ contention that Lindblad fails to identify semantic content of the upper overlapping sub-tree structure as relevant to the classifying step (App. Br. 16) is not only not commensurate with the scope of the claim, it fails to persuasively rebut the Examiner’s findings based on Lindblad’s Paragraph 0042 and Figure 3 (Ans. 6) and for the reasons noted above. Claim 3 We also sustain the Examiner’s rejection of claim 3 reciting concatenating the fragment’s content components. Appellants’ contention regarding the concatenation of the outer tree structure providing a more robust recognition (App. Br. 16-17) fails to persuasively rebut the Examiner’s findings based on Lindblad’s Paragraph 0042 and Figure 3 (Ans. 6-7) and for the reasons noted above. Claim 4 We also sustain the Examiner’s rejection of claim 4 reciting recognizing a structural element of the fragment as the predetermined structure. Appellants’ contention regarding recognizing an outer tree Appeal 2010-007140 Application 11/156,776 9 structure as having a predetermined structure (App. Br. 17) fails to persuasively rebut the Examiner’s findings based on Lindblad’s Paragraph 0063 (Ans. 7) and for the reasons noted above. Claim 5 We also sustain the Examiner’s rejection of claim 5 reciting identifying the fragment’s relative position within the document. Appellants’ contention that Lindblad is not concerned with outer structure (App. Br. 17) fails to persuasively rebut the Examiner’s findings based on Lindblad’s Paragraph 0063, Table 1, and Figure 14 (Ans. 7) and for the reasons noted above. Claim 6 We also sustain the Examiner’s rejection of claim 6 reciting identifying a logical structure of the fragment. Appellants’ contention that Lindblad fails to recognize the outer logical structure (App. Br. 17) is not commensurate with the scope of the claim and also fails to persuasively rebut the Examiner’s findings based on Lindblad’s Paragraph 0063, Table 1, and Figure 14 (Ans. 7) and for the reasons noted above. Claim 7 We also sustain the Examiner’s rejection of claim 7 reciting (a) representing the fragment as a sub-tree having a navigational path between a fragment root and leaf, and (b) defining the logical structure as the navigational path. Appellants’ contention regarding identifying the outer structure’s navigational path (App. Br. 17) is not commensurate with the Appeal 2010-007140 Application 11/156,776 10 scope of the claim and also fails to persuasively rebut the Examiner’s findings based on Lindblad’s Paragraphs 0043 and 0044 and Figures 4A and 4B (Ans. 7-8) and for the reasons noted above. Claims 12 and 19 We also sustain the Examiner’s rejection of claim 12 reciting determining distinct likelihoods of correspondence to the type of document for the sub-tree outer and inner structures. Appellants’ contention that the distinct likelihoods of correspondence using the classifying is limited to include both outer and inner sub-tree structures (App. Br. 18) fails to persuasively rebut the Examiner’s findings based on Lindblad’s Paragraph 0065 and Figure 14 (Ans. 8) and for the reasons noted above. We reach a similar conclusion regarding claim 19 which recites commensurate limitations, because Appellants’ arguments (App. Br. 19) do not persuasively rebut the Examiner’s findings (Ans. 10) regarding this claim. Claim 13 We also sustain the Examiner’s rejection of claim 13 reciting combining the distinct likelihoods for estimating a final document type. Appellants’ contention regarding combining the likelihoods of both inner and outer structures (App. Br. 18) fails to persuasively rebut the Examiner’s findings based on Lindblad’s Paragraph 0065 (Ans. 8-9) and for the reasons noted above. Appeal 2010-007140 Application 11/156,776 11 Claim 14 We also sustain the Examiner’s rejection of claim 14 reciting that the likelihoods are weighted by a pre-selected weight. Appellants’ contention regarding weighting the likelihoods of both inner and outer structures (App. Br. 18) fails to persuasively rebut the Examiner’s findings based on Lindblad’s Paragraph 0065 and Figure 14 (Ans. 9) and for the reasons noted above. Claims 16 and 20 We also sustain the Examiner’s rejection of claim 16 reciting determining distinct likelihoods of correspondence to the type of document for each of the sub-tree outer structure, sub-tree inner structure, sub-tree content, and sub-tree node tags and attributes. Appellants’ contention regarding determining the likelihoods of correspondence with respect to the outer structure and other recited elements (App. Br. 18-19) fails to persuasively rebut the Examiner’s findings based on Lindblad’s Paragraph 0063 (Ans. 9) and for the reasons noted above. We reach a similar conclusion regarding claim 20 which recites commensurate limitations, because Appellants’ arguments (App. Br. 19) do not persuasively rebut the Examiner’s findings (Ans. 10) regarding this claim. Claim 17 We also sustain the Examiner’s rejection of claim 17 reciting combining the distinct likelihood [sic] for estimating a final document type. Appellants’ contention that Lindblad fails to combine distinct likelihoods of inner and outer structures to estimate a final document type (App. Br. 19) Appeal 2010-007140 Application 11/156,776 12 fails to persuasively rebut the Examiner’s findings based on Lindblad’s Paragraph 0065 (Ans. 9) and for the reasons noted above. THE OBVIOUSNESS REJECTION The Examiner finds that Lindblad assigns a selected class for a document on a basis of each one of the selectively identified plurality of predetermined contents and structures, but does not weigh the assigned selected classes and determine a final class from combining the weighted classes. Ans. 11, 16. The Examiner, however, cites Zha as teaching these features in concluding that the claim would have been obvious. Ans. 11-12, 16-17. Appellants argue that while Zha applies weighted classifications, the classifications are not applied to an outer tree structure when determining attributes and contents of a particular inner tree structure to the relevant outer structure. App. Br. 20. ISSUE Under § 103, has the Examiner erred in rejecting claim 9 by finding that Lindblad and Zha, collectively, would have taught or suggested classifying comprises (1) assigning a selected class for the document on a basis of each one of the selectively identified plurality of the predetermined content and structure, (2) weighting the assigned selected classes, and (3) determining a final class from combining the weighted classes? Appeal 2010-007140 Application 11/156,776 13 ANALYSIS On this record, we find no error in the Examiner’s obviousness rejection of representative claim 9. At the outset, we note that Appellants do not squarely rebut the Examiner’s findings regarding determining a final class from combining weighted classes, but rather contest the other limitations of the claim as they pertain to the recited classification process. See App. Br. 20; Reply Br. 6. Nevertheless, Appellants’ arguments (App. Br. 20) regarding Zha’s alleged shortcomings concerning applying weighted classifications to an outer tree structure for classification do not persuasively rebut the Examiner’s reliance on Lindblad’s and Zha’s collective teachings. As noted above, the Examiner relies on Lindblad for teaching the class assigning step, and Zha for the weighting and determining steps. Ans. 11- 12, 16-17. As previously discussed, we find no error in the Examiner’s reliance on Lindblad for teaching the recited classification based on relative document similarity which is based partly on the document’s outer structure. Nor are we persuaded of error in the Examiner’s reliance on Zha which, as Appellants acknowledge (App. Br. 20), applies weighted classifications (i.e., region and language) to rank documents accordingly. Zha col. 3, l. 5 – col. 5, l. 20; Figs. 1-2C. Applying this teaching to Lindblad would predictably yield weighted document similarity classifications or classes based, at least partly, on the tree’s and sub-tree’s content and structure. To the extent that Appellants contend that these classifications are not classes,3 we find such arguments unavailing. The weight of the evidence on this record, therefore, favors the Examiner’s position. 3 See Reply Br. 6 (asserting that neither cited reference teaches classes). Appeal 2010-007140 Application 11/156,776 14 We are therefore not persuaded that the Examiner erred in rejecting representative claim 9, and claim 10 not separately argued with particularity. CONCLUSION The Examiner did not err in rejecting (1) claims 1-8 and 11-20 under § 102, and (2) claims 9 and 10 under § 103. ORDER The Examiner’s decision rejecting claims 1-20 is affirmed. 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)(iv). AFFIRMED babc Copy with citationCopy as parenthetical citation