Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Specification
The disclosure is objected to because of the following informalities: the table of contents at ¶[14] recites section titles “2. METAVERSE ENVIRONEMENT SELECTION ARCHITECTURE” and “3. METAVERSE ENVIRONEMENT SELECTION SYSTEM,” which misspell “ENVIRONMENT” as “ENVIRONEMENT” and which do not correspond to the actual section headings appearing at ¶[20] (“2. ENVIRONMENT SELECTION ARCHITECTURE”) and ¶[26] (“3. ENVIRONMENT SELECTION SYSTEM”); similarly, the table of contents entries “4. SELECTING METAVERSE ENVIRONMENTS FOR CONTENT” and “5. EXAMPLE EMBODIMENT OF SELECTING METAVERSE ENVIRONMENTS FOR CONTENT” do not correspond to the headings at ¶[43] (“4. SELECTING SUITABLE ENVIRONMENTS FOR PLACING CONTENT”) and ¶[63] (“5. EXAMPLE EMBODIMENT OF ENVIRONMENT SELECTION”); At ¶[33] refers to “environments 112,” inconsistent with reference character 110 (110A-110C), which is used to designate the environments throughout the remainder of the specification and in FIGS. 1A and 3; At ¶[34] and a subsequent paragraph refer to element 106 as the “content placement system,” inconsistent with the “environment selection system 106” terminology used elsewhere in the specification and in FIGS. 1A, 1B, and 3; At ¶[36] contains the duplicated word “the the” (“...populates the the compatibility score database 156...”); At ¶[38] refers to “training database 150,” inconsistent with reference character 140, which designates the training database elsewhere in the specification (¶[28]) and in FIG. 1B; and at ¶[42] refers to “an environment 100,” inconsistent with reference character 110, which designates an environment, as reference character 100 designates the environment selection architecture as a whole (FIG. 1A).
Appropriate correction is required.
Claim Objections
Claims 5, 12, and 19 are objected to because of the following informalities: each claim recites “...based on or more of: metadata...”; the phrase should read “based on one or more of.”
Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 1 recites a colon-introduced sub-list defining what "individual training sets of the plurality of training sets" comprise (i.e., "attributes of a particular content item; attributes of a particular metaverse environment; a particular compatibility score..."), immediately followed — using the identical semicolon punctuation and without any closing terminator (e.g., "; and") demarcating the end of that sub-list — by the further recitations "training the machine learning model based on the plurality of training sets; identifying a first content item; identifying a first candidate metaverse environment...; applying the machine learning model...; and ... selecting the first candidate metaverse environment...." Because the sub-list (defining the contents of a training set) and the outer list (defining the operations performed by the claimed instructions) are both introduced by a colon and separated using identical semicolon punctuation with no distinguishing indentation, numbering, or closing language, the claim does not distinctly indicate where the inner "comprising:" list defining the training-set contents ends and the outer "operations comprising:" list resumes. A person of ordinary skill in the art cannot determine with reasonable certainty, from the claim language and punctuation alone, the full boundary and membership of each list, rendering the metes and bounds of the claim uncertain.
For purposes of examination, the claim is interpreted under BRI such that the sub-list introduced by "individual training sets of the plurality of training sets comprising:" is limited to the three data-content items expressly reciting "attributes of a particular content item," "attributes of a particular metaverse environment," and "a particular compatibility score," and that "training the machine learning model based on the plurality of training sets," "identifying a first content item," "identifying a first candidate metaverse environment," "applying the machine learning model," and "selecting the first candidate metaverse environment" are each construed as separate top-level operations of the outer "operations comprising:" list.
Claim 8 recites the identical ambiguously nested "comprising:" list structure discussed above with respect to claim 1, applied to a method claim. The sub-list defining the contents of "individual training sets" is not distinctly demarcated from the subsequently recited method steps ("training the machine learning model...," "identifying a first content item," etc.), which are introduced and separated using the same punctuation as the sub-list. The claim therefore fails to distinctly point out the boundary between the recited data elements and the recited method steps.
For purposes of examination, claim 8 is interpreted under BRI in the same manner as claim 1 above: the sub-list is limited to the three data-content items, and "training the machine learning model," "identifying a first content item," "identifying a first candidate metaverse environment," "applying the machine learning model," and "selecting the first candidate metaverse environment" are construed as separate method steps.
Claim 15 recites the identical ambiguously nested "comprising:" list structure discussed above with respect to claim 1, applied to a system claim reciting operations the system is "configured to perform." As with claims 1 and 8, the sub-list defining the contents of "individual training sets" is not distinctly demarcated from the subsequently recited operations, which are introduced and separated using the same punctuation as the sub-list.
For purposes of examination, claim 15 is interpreted under BRI in the same manner as claims 1 and 8 above: the sub-list is limited to the three data-content items, and the remaining recitations are construed as separate operations the system is configured to perform.
Claims 2-7 depend, directly or indirectly, from claim 1 and incorporate all of the limitations of claim 1, including the ambiguously nested "comprising:" list structure identified above. Claims 2-7 do not resolve the ambiguity and are therefore indefinite for at least the same reason as claim 1.
Claims 9-14 depend, directly or indirectly, from claim 8 and incorporate all of the limitations of claim 8, including the ambiguously nested "comprising:" list structure identified above. Claims 9-14 do not resolve the ambiguity and are therefore indefinite for at least the same reason as claim 8.
Claims 16-20 depend, directly or indirectly, from claim 15 and incorporate all of the limitations of claim 15, including the ambiguously nested "comprising:" list structure identified above. Claims 16-20 do not resolve the ambiguity and are therefore indefinite for at least the same reason as claim 15.
Claim 5 recites "based on or more of" in place of the presumably intended "based on one or more of." As written, the phrase "based on or more of" is grammatically incomplete and does not convey a clear, ascertainable meaning. A person of ordinary skill in the art cannot determine with reasonable certainty whether the claim requires identification based on a single listed item, any combination of the listed items, or all of the listed items, rendering the scope of the limitation indefinite.
For purposes of examination, "based on or more of" is interpreted under BRI to mean "based on one or more of," consistent with the evident intent of the sentence and the specification at ¶[39], which describes the attribute generation module extracting keywords from metadata, code, and scraped content, and identifying objects within an environment — i.e., from any one or combination of these sources.
Claim 12 recites the identical incomplete phrase "based on or more of" discussed above with respect to claim 5, for at least the same reasons.
Claim 19 recites the identical incomplete phrase "based on or more of" discussed above with respect to claims 5 and 12, for at least the same reasons.
Appropriate correction is required.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
CLAIM 1
Step 1: Claim 1 recites "[o]ne or more non-transitory computer readable media comprising instructions," which is an article of manufacture. See MPEP 2106.03. Accordingly, claim 1 satisfies Step 1 of the eligibility analysis.
Step 2A, Prong 1: Claim 1 is directed to an abstract idea. Specifically, claim 1 recites (i) a certain method of organizing human activity — a commercial interaction in the nature of advertising or marketing, namely selecting an environment in which to place promotional content to reach a target audience — and, in the alternative, (ii) a mental process — the human evaluation or judgment of how compatible a content item is with a candidate environment. The following limitations recite the abstract idea: “compute a first compatibility score representing a first level of compatibility between the first content item and the first candidate metaverse environment” (understood as arriving at a compatibility judgment) and “based at least on the first compatibility score, selecting the first candidate metaverse environment as a target metaverse environment for placement of the first content item”. Under their broadest reasonable interpretation, forming a judgment of how compatible a content item is with an environment, and selecting an environment for placement of the content based on that judgment, encompass evaluations, judgments, and opinions that can practically be performed in the human mind and constitute the commercial act of targeting advertising. The specification confirms that the compatibility determination is an ordinary human evaluation, explaining that “an advertisement for beer may be incompatible with a children’s farming simulation and compatible with a sporting event simulation” (spec ¶[16]) and that compatibility scores can be “assigned by subject matter experts” (spec ¶[29]). See MPEP 2106.04(a)(2), subsection II (certain methods of organizing human activity) and subsection III (mental processes).
Step 2A, Prong 2: The additional elements are: “[o]ne or more non-transitory computer readable media comprising instructions that, when executed by one or more hardware processors, causes performance of operations”; “identifying a first content item” and “identifying a first candidate metaverse environment for the first content item” (data gathering); “obtaining a plurality of training sets for training a machine learning model” and the recited training-set data; “training the machine learning model based on the plurality of training sets”; and “applying the machine learning model to attributes of the first content item and attributes of the first candidate metaverse environment”.
The judicial exception is not integrated into a practical application. The additional elements, considered individually and as an ordered combination, do not integrate the abstract idea into a practical application.
Under Ex Parte Desjardins, Appeal No. 2024-000567 (PTAB Sept. 26, 2025), the specification is first evaluated to determine whether it describes an improvement to the functioning of a computer or to another technology, and, if so, whether the claim reflects that improvement. Here, the specification does not describe any improvement to computer functionality or to machine learning technology itself. The specification employs conventional, off-the-shelf machine learning, stating that the model may use any of "linear regression, logistic regression, linear discriminant analysis, classification and regression trees, naïve Bayes, k-nearest neighbors, learning vector quantization, support vector machine, bagging and random forest, boosting, backpropagation, and/or clustering" (spec ¶[30]), and that training is ordinary supervised learning in which "the algorithm iteratively learns the relationship between the inputs and labels" (spec ¶[48]).
Unlike the claims in Ex Parte Desjardins — which reflected a specific improvement to how a machine learning model operates (training a model to learn new tasks while protecting knowledge of prior tasks to overcome "catastrophic forgetting," reducing storage requirements and system complexity) — the present specification describes using a generic machine learning model merely as a tool to compute a compatibility score and to automate a content-placement (advertising) decision. The asserted benefit, placing content in a more compatible environment, is an improvement to the abstract idea (targeted content placement) itself, not an improvement to technology. See MPEP 2106.05(a); Recentive Analytics, Inc. v. Fox Corp., 134 F.4th 1205 (Fed. Cir. 2025) (applying generic machine learning to automate an abstract idea does not confer eligibility). Because the specification does not disclose a technological improvement, the first step of the Desjardins analysis is not satisfied, and the claim does not reflect any such improvement.
The recited non-transitory computer readable media and hardware processors are set forth at a high level of generality and amount to no more than mere instructions to apply the abstract idea using generic computer components. See MPEP 2106.05(f). The specification confirms the generic nature of these components, describing a "general purpose microprocessor" (spec ¶[74]) and a conventional computer system (spec ¶¶[73]-[85], FIG. 4). The "machine learning model", "training the machine learning model", and "applying the machine learning model" limitations use a machine learning model merely as a tool to perform the abstract evaluation and thus amount to "apply it" on a computer. See MPEP 2106.05(f); Recentive Analytics, Inc. v. Fox Corp., 134 F.4th 1205 (Fed. Cir. 2025).
"[O]btaining a plurality of training sets" and "identifying a first content item" / "identifying a first candidate metaverse environment" are insignificant extra-solution activity in the nature of mere data gathering. See MPEP 2106.05(g). The recitation of a "metaverse" environment merely confines the abstract idea to a particular technological environment or field of use. See MPEP 2106.05(h). The claim recites no metaverse-specific technological operation — e.g., no rendering, spatial placement, or real-time interaction within the virtual environment — but only the computation of a compatibility score and the selection of an environment; the claim therefore does not effect a technological solution rooted in the computer environment in the manner of DDR Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245 (Fed. Cir. 2014), but merely applies the abstract idea within that environment. Considered as an ordered combination, these additional elements add nothing that integrates the abstract idea into a practical application. Accordingly, claim 1 does not satisfy Step 2A, Prong 2.
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception. Considered individually and as an ordered combination, the additional elements are well-understood, routine, and conventional. The recited non-transitory computer readable media and hardware processors that store and execute instructions are well-understood, routine, and conventional generic computer components. See Alice Corp. v. CLS Bank Int'l, 573 U.S. 208, 226 (2014); MPEP 2106.05(d). The specification's own description of a "general purpose microprocessor" (spec ¶[74]) and a conventional computer system (spec ¶¶[73]-[85]) is evidence that these components are conventional. Using a machine learning model as a tool to perform repetitive calculations and automate the abstract idea is not significantly more, as the specification describes the model in terms of conventional, named off-the-shelf algorithms and ordinary supervised training (spec ¶[30], ¶[48]). See Recentive Analytics, Inc. v. Fox Corp., 134 F.4th 1205 (Fed. Cir. 2025). "[O]btaining a plurality of training sets" and "identifying" a content item and a candidate environment are well-understood, routine, and conventional data gathering. See Electric Power Group, LLC v. Alstom S.A., 830 F.3d 1350, 1355 (Fed. Cir. 2016); MPEP 2106.05(d)(II) and 2106.05(g). The ordered combination adds nothing beyond the individual elements because each element performs only its ordinary and expected function — the media store and the processor executes the instructions, and the machine learning model computes a score — with no unconventional interaction among the elements that could supply an inventive concept. See BASCOM Global Internet Servs. v. AT&T Mobility LLC, 827 F.3d 1341, 1350 (Fed. Cir. 2016). The ordered combination of these elements adds nothing beyond the sum of the individual conventional elements. Accordingly, claim 1 does not satisfy Step 2B and is rejected under 35 U.S.C. 101.
CLAIM 2
Step 1: Claim 2 depends from claim 1 and recites one or more non-transitory computer readable media, which falls within the statutory category of a manufacture. See MPEP 2106.03.
Step 2A, Prong 1: Claim 2 additionally recites "identifying a second candidate metaverse environment", "compute a second compatibility score representing a second level of compatibility between the first content item and the second candidate metaverse environment", and "determining that the first compatibility score is higher than the second compatibility score", whereby the first candidate environment is selected based on that comparison. These are further evaluation, comparison, and selection steps of the same mental-process nature as those in claim 1 — a person can evaluate the compatibility of a second environment, compare the two compatibility scores, and select the higher-scoring environment mentally. See MPEP 2106.04(a)(2).
Step 2A, Prong 2: The additional elements are "applying the machine learning model…" to the second environment. The judicial exception is not integrated into a practical application. Claim 2 introduces no new type of additional element beyond the machine learning tool already analyzed for claim 1; the further "identifying," score-computing, "determining," and comparison-based selection limitations are themselves part of the abstract idea. For the same reasons discussed in the Step 2A, Prong 2 analysis of claim 1 — including that the specification discloses no improvement to computer functionality or to machine learning technology under Ex Parte Desjardins — the additional elements do not integrate the judicial exception into a practical application. Accordingly, claim 2 does not satisfy Step 2A, Prong 2.
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception. Claim 2 introduces no new types of additional element; the further application of the machine learning model to a second environment is the same conventional use of a machine learning tool as in claim 1. For the same reasons discussed in the Step 2B analysis of claim 1, the additional elements, individually and as an ordered combination, are well-understood, routine, and conventional and do not amount to significantly more. Accordingly, claim 2 does not satisfy Step 2B and is rejected under 35 U.S.C. 101.
CLAIM 3
Step 1: Claim 3 depends from claim 1 and recites one or more non-transitory computer readable media, which falls within the statutory category of a manufacture. See MPEP 2106.03.
Step 2A, Prong 1: There are no additional abstract ideal limitations introduced.
Step 2A, Prong 2: Claim 3 additionally recites that "applying the machine learning model comprises: computing an environment feature vector based on keywords associated with the first candidate metaverse environment". The "keywords associated with the first candidate metaverse environment" are observations of attributes that a person can identify mentally and are part of the abstract evaluation of context. The recited "computing an environment feature vector based on keywords associated with the first candidate metaverse environment" is a data-representation step performed as part of applying the machine learning model; it does not set forth or name any particular mathematical formula or algorithm and is treated, together with the machine learning model, as an additional element. See MPEP 2106.04(a)(2). No new abstract-idea concept beyond the compatibility evaluation of claim 1 is introduced.
The judicial exception is not integrated into a practical application. The additional element of "computing an environment feature vector based on keywords associated with the first candidate metaverse environment" is part of applying the machine learning model and represents, at most, the selection and numerical representation of a particular type of input data for the machine learning tool — insignificant extra-solution activity (data gathering and formatting) that does not integrate the abstract idea (see MPEP 2106.05(g)). For the same reasons discussed in the Step 2A, Prong 2 analysis of claim 1 — including that the specification discloses no improvement to computer functionality or to machine learning technology under Ex Parte Desjardins — the additional elements do not integrate the judicial exception into a practical application. Accordingly, claim 3 does not satisfy Step 2A, Prong 2.
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception. Computing an environment feature vector from keywords is a routine data-formatting and repetitive calculation operation performed by the machine learning tool and is well-understood, routine, and conventional, as reflected in the specification's generic description of feature vectors as "one-dimensional arrays containing attributes" applied as inputs to conventional algorithms (spec ¶[35], ¶[30]). See Electric Power Group, LLC v. Alstom S.A., 830 F.3d 1350, 1355 (Fed. Cir. 2016); MPEP 2106.05(d)(II). For the same reasons discussed in the Step 2B analysis of claim 1, the additional elements, individually and as an ordered combination, are well-understood, routine, and conventional and do not amount to significantly more. Accordingly, claim 3 does not satisfy Step 2B and is rejected under 35 U.S.C. 101.
CLAIM 4
Step 1: Claim 4 depends from claim 3 and recites one or more non-transitory computer readable media, which falls within the statutory category of a manufacture. See MPEP 2106.03.
Step 2A, Prong 1: Claim 4 additionally recites "determining the keywords associated with the first candidate metaverse environment by scraping information from the first candidate metaverse environment". Per "determining the keywords associated with the first candidate metaverse environment", this is an observation and identification step of the same mental-process nature as the evaluation in claim 1.
Step 2A, Prong 2: Claim 4 recites the additional elements”: "scraping information from the first candidate metaverse environment" — is insignificant extra-solution activity in the nature of mere data gathering (collecting the keyword data used in the abstract evaluation), and imposes no meaningful limit on the abstract idea (see MPEP 2106.05(g)). For the same reasons discussed in the Step 2A, Prong 2 analysis of claim 1 — including that the specification discloses no improvement to computer functionality or to machine learning technology under Ex Parte Desjardins — the additional elements do not integrate the judicial exception into a practical application. Accordingly, claim 4 does not satisfy Step 2A, Prong 2.
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception. Scraping (i.e., electronically collecting and extracting) information from a data source is well-understood, routine, and conventional. See Content Extraction & Transmission LLC v. Wells Fargo Bank, N.A., 776 F.3d 1343, 1348 (Fed. Cir. 2014); MPEP 2106.05(d)(II). The specification confirms that such scraping uses conventional techniques such as "natural language processing (NLP), image recognition, and semantic analysis" (spec ¶¶[33]-[34], ¶[39]). For the same reasons discussed in the Step 2B analysis of claim 1, the additional elements, individually and as an ordered combination, are well-understood, routine, and conventional and do not amount to significantly more. Accordingly, claim 4 does not satisfy Step 2B and is rejected under 35 U.S.C. 101.
CLAIM 5
Step 1: Claim 5 depends from claim 4 and recites one or more non-transitory computer readable media, which falls within the statutory category of a manufacture. See MPEP 2106.03.
Step 2A, Prong 1: Claim 5 additionally recites "identifying the keywords associated with the first candidate metaverse environment based on [one] or more of: metadata associated with the first candidate metaverse environment; code associated with the first candidate metaverse environment; subject matter displayed in the first candidate metaverse environment; and objects included in the first candidate metaverse environment". "[I]dentifying the keywords" is an observation/evaluation step of the same mental-process nature as claim 1; the recited metadata, code, subject matter, and objects merely specify the particular source or type of data from which the keywords are identified. No new abstract-idea concept and no new type of additional element is introduced. See MPEP 2106.04(a)(2).
Step 2A, Prong 2: The judicial exception is not integrated into a practical application. Specifying the particular source or type of data (metadata, code, displayed subject matter, or objects) from which keywords are identified is the selection of a particular data source, which can be construed as insignificant extra-solution activity — and further merely confines the abstract idea to a particular field of use (see MPEP 2106.05(g) and (h)). For the same reasons discussed in the Step 2A, Prong 2 analysis of claim 1 — including that the specification discloses no improvement to computer functionality or to machine learning technology under Ex Parte Desjardins — the additional elements do not integrate the judicial exception into a practical application. Accordingly, claim 5 does not satisfy Step 2A, Prong 2.
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception. Selecting a particular data source and extracting keyword data from metadata, code, displayed subject matter, or objects is well-understood, routine, and conventional data gathering. See Content Extraction & Transmission LLC v. Wells Fargo Bank, N.A., 776 F.3d 1343, 1348 (Fed. Cir. 2014); Electric Power Group, LLC v. Alstom S.A., 830 F.3d 1350, 1355 (Fed. Cir. 2016); MPEP 2106.05(d)(II) and 2106.05(g). For the same reasons discussed in the Step 2B analysis of claim 1, the additional elements, individually and as an ordered combination, are well-understood, routine, and conventional and do not amount to significantly more. Accordingly, claim 5 does not satisfy Step 2B and is rejected under 35 U.S.C. 101.
CLAIM 6
Step 1: Claim 6 depends from claim 1 and recites one or more non-transitory computer readable media, which falls within the statutory category of a manufacture. See MPEP 2106.03.
Step 2A, Prong 1: Claim 6 additionally recites that "applying the machine learning model comprises: computing a content feature vector based on keywords associated with the first content item". The "keywords associated with the first content item" are observations of attributes that a person can identify mentally and are part of the abstract evaluation of context. No new abstract-idea concept beyond the compatibility evaluation of claim 1 is introduced.
Step 2A, Prong 2: The claim additionally recites "computing a content feature vector based on keywords associated with the first content item", this being a data-representation step performed as part of applying the machine learning model; it does not set forth or name any particular mathematical formula or algorithm and is treated, together with the machine learning model, as an additional element. See MPEP 2106.04(a)(2).
The judicial exception is not integrated into a practical application. The additional element of "computing a content feature vector based on keywords associated with the first content item" is part of applying the machine learning model and represents, at most, the selection and numerical representation of a particular type of input data for the machine learning tool — insignificant extra-solution activity (data gathering and formatting) that does not integrate the abstract idea (see MPEP 2106.05(g)). For the same reasons discussed in the Step 2A, Prong 2 analysis of claim 1 — including that the specification discloses no improvement to computer functionality or to machine learning technology under Ex Parte Desjardins — the additional elements do not integrate the judicial exception into a practical application. Accordingly, claim 6 does not satisfy Step 2A, Prong 2.
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception. Computing a content feature vector from keywords is a routine data-formatting and repetitive calculation operation performed by the machine learning tool and is well-understood, routine, and conventional, as reflected in the specification's generic description of feature vectors as "one-dimensional arrays containing attributes" applied as inputs to conventional algorithms (spec ¶[35], ¶[30]). See Electric Power Group, LLC v. Alstom S.A., 830 F.3d 1350, 1355 (Fed. Cir. 2016); MPEP 2106.05(d)(II). For the same reasons discussed in the Step 2B analysis of claim 1, the additional elements, individually and as an ordered combination, are well-understood, routine, and conventional and do not amount to significantly more. Accordingly, claim 6 does not satisfy Step 2B and is rejected under 35 U.S.C. 101.
CLAIM 7
Step 1: Claim 7 depends from claim 6 and recites one or more non-transitory computer readable media, which falls within the statutory category of a manufacture. See MPEP 2106.03.
Step 2A, Prong 1: Claim 7 additionally recites "determining the keywords associated with the first content item by scraping information from the first content item", where "determining the keywords associated with the first content item" is an observation and identification step of the same mental-process nature as the evaluation in claim 1.
Step 2A, Prong 2: The recites the new additional element of "scraping information from the first content item". This is insignificant extra-solution activity in the nature of mere data gathering (collecting the keyword data used in the abstract evaluation), and imposes no meaningful limit on the abstract idea (see MPEP 2106.05(g)). For the same reasons discussed in the Step 2A, Prong 2 analysis of claim 1 — including that the specification discloses no improvement to computer functionality or to machine learning technology under Ex Parte Desjardins — the additional elements do not integrate the judicial exception into a practical application. Accordingly, claim 7 does not satisfy Step 2A, Prong 2.
Step 2B: The claim does not include additional elements that amount to significantly more than the judicial exception. Scraping (i.e., electronically collecting and extracting) information from a data source is well-understood, routine, and conventional. See Content Extraction & Transmission LLC v. Wells Fargo Bank, N.A., 776 F.3d 1343, 1348 (Fed. Cir. 2014); MPEP 2106.05(d)(II). The specification confirms that such scraping uses conventional techniques such as "natural language processing (NLP), image recognition, and semantic analysis" (spec ¶¶[33]-[34], ¶[39]). For the same reasons discussed in the Step 2B analysis of claim 1, the additional elements, individually and as an ordered combination, are well-understood, routine, and conventional and do not amount to significantly more. Accordingly, claim 7 does not satisfy Step 2B and is rejected under 35 U.S.C. 101.
Claims 8-14 are substantially similar in scope and spirit as claims 1-7, therefore, the rejection of claim 1-7 are applied accordingly. The main difference for claims 8-14, is the Step 1 analysis where claims 8-14 recite "[a] method," which is a series of steps or acts. See MPEP 2106.03. Accordingly, claims 8-14 satisfies Step 1 of the eligibility analysis.
Claims 15-20 are substantially similar in scope and spirit as claims 1-7, therefore, the rejection of claim 1-7 are applied accordingly. The main difference for claims 15-20, is the Step 1 analysis where claims 15-20 recite "[a] system comprising: at least one device including a hardware processor," which is a machine. See MPEP 2106.03. Accordingly, claim 15 satisfies Step 1 of the eligibility analysis.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over US Pat. Pub. No. 2022/0245901A1 to Goyal et al. (hereinafter Goyal) in view of US Pat. Pub. No. 2012/0259712A1 to Hyndman et al. (hereinafter Hyndman).
Per claim 1, Goyal discloses One or more non-transitory computer readable media comprising instructions that, when executed by one or more hardware processors, causes performance of operations (Goyal: FIG. 5:504 and ¶[0028]…Goyal's control circuitry executes the disclosed operations out of storage, and a PHOSITA would understand such control circuitry to comprise one or more hardware processors executing instructions held on non-transitory media within the BRI of the preamble, "processing circuitry should be understood to mean circuitry based on one or more microprocessors, microcontrollers, digital signal processors, programmable logic devices, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), etc."), comprising:
obtaining a plurality of training sets for training a machine learning model, individual training sets of the plurality of training sets comprising (Goyal: FIG. 1 and ¶[0023]…Goyal assembles training sets from lists of candidate, persistent, and temporary virtual objects used to train the model, which reads on obtaining a plurality of training sets under BRI, "each temporary object from temporary object list 128 may be used as a training candidate object, persistent object list 126 may be used as the set of persistent virtual objects, and the remaining objects from temporary object list 128 may be used as the set of temporary virtual objects"):
attributes of a particular content item (Goyal: FIG. 3:302 and ¶[0025]… each candidate/temporary virtual object (the content item to be inserted) carries object attributes such as type, class, family, genus, and species, which constitute attributes of a particular content item under BRI, "sun 108 is of type 'persistent,' meaning it is a persistent object. Sun 108 is of class 'natural,'... Genus and species attributes further define sun 108 as a 'star' and a 'local sun,' respectively");
attributes of a particular metaverse environment (Goyal: FIG. 1, FIG. 3 and ¶[0023]…the persistent and temporary virtual objects displayed in a given virtual environment, together with their object attributes, characterize that environment and are supplied to the model, which constitutes attributes of a particular metaverse environment under BRI, "persistent object list 126 may be used as the set of persistent virtual objects, and the remaining objects from temporary object list 128 may be used as the set of temporary virtual objects");
a particular compatibility score representing a particular level of compatibility between the particular content item and the particular metaverse environment (Goyal: FIG. 2:206 and ¶[0024]…Goyal's model is a gradient boosted decision tree for classification trained on training data set 130, so each training example is labeled by whether the candidate object fits the environment defined by the persistent and temporary objects, and the confidence value the model is trained to output is that per-pair level of compatibility, which constitutes a particular compatibility score under BRI, "Trained machine learning model 202 outputs confidence level 206 that candidate object 204 fits into the virtual environment"; FIG. 1 and ¶[0023]…"The machine learning model is then trained with a target of returning a confidence prediction for each candidate object to be near 100%");
training the machine learning model based on the plurality of training sets (Goyal: FIG. 1 and ¶[0023]…Goyal trains the machine learning model on the assembled sets of candidate, persistent, and temporary objects, which constitutes training the model based on the plurality of training sets, "The machine learning model is then trained with a target of returning a confidence prediction for each candidate object to be near 100%");
identifying a first content item (Goyal: FIG. 5:520 and ¶[0038]…Goyal identifies a candidate (supplemental content) object to be evaluated by the trained model, which constitutes identifying a first content item under BRI, "Machine learning model 520 calculates a confidence for each input candidate object");
applying the machine learning model to attributes of the first content item and attributes of the first candidate metaverse environment to compute a first compatibility score representing a first level of compatibility between the first content item and the first candidate metaverse environment (Goyal: FIG. 2:202,204,206 and ¶[0024]…the trained model receives the candidate object's attributes and the environment's attributes and outputs a confidence level that the candidate object fits the environment, which is the claimed first compatibility score under BRI, "Trained machine learning model 202 outputs confidence level 206 that candidate object 204 fits into the virtual environment"; FIG. 2:206 and ¶[0024] the score quantifies the level of compatibility, "A high confidence level indicates that candidate object 204 fits into the virtual environment and can thus be inserted into the virtual environment at a suitable location").
Goyal does not expressly disclose, but with Hyndman does teach:
identifying a first candidate metaverse environment for the first content item (Hyndman: Summary…Hyndman selects, for a given item of advertising content, a predefined space of the three-dimensional virtual environment to receive that content, which constitutes identifying a first candidate metaverse environment for the first content item under BRI, "A predefined space is selected for receiving the advertising content from an advertising provider");
based at least on the first compatibility score, selecting the first candidate metaverse environment as a target metaverse environment for placement of the first content item (Hyndman: Summary…Hyndman selects the candidate space to receive the content based on the context of that space, supplying the environment-as-target that Goyal's object-centric selection does not expressly recite, and in the combination the selection is governed by Goyal's computed compatibility score, which Goyal uses to place content, "The candidate object having the highest confidence is then selected and transmitted 532 to virtual environment generation circuitry 506, which incorporates the selected candidate object into the virtual environment" (Goyal: FIG. 5:520,532 and ¶[0038]), so a PHOSITA selects, for the content item, the candidate environment on the basis of the first compatibility score, "Information regarding the context of the space within the virtual environment is used by the advertising provider to select appropriate advertising content that relates to the context of the space within the virtual environment where the advertising content is to be placed").
Goyal and Hyndman are analogous art because they are from the same field of endeavor, specifically the automated placement of content items - advertisements or other supplemental virtual objects - into three-dimensional virtual (metaverse) environments. They address the same problem of matching a content item to a compatible environment for placement of the content.
Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art (PHOSITA) to apply the trained machine-learning compatibility-scoring model of Goyal to select, for a given content item, the candidate three-dimensional virtual environment of Hyndman whose context is most compatible with that content item.
The suggestion/motivation for doing so is provided by Hyndman itself, which teaches selecting a predefined space to receive advertising content based on the context of that space so that "the appropriate content can be delivered by recognizing the key words" (Hyndman: FIG. 2:S32 and ¶31). A PHOSITA would have been motivated to compute Goyal's compatibility (confidence) score for the candidate environment in order to automate and improve the accuracy of Hyndman's context-based selection of an environment for a content item, yielding the predictable result of placing the content item in its most compatible environment (KSR rationale (A) - combining prior art elements according to known methods to yield predictable results).
Per claim 2, Goyal combined with Hyndman discloses claim 1. Goyal with Hyndman further teaches identifying a second candidate metaverse environment; applying the machine learning model to the attributes of the first content item and attributes of the second candidate metaverse environment to compute a second compatibility score representing a second level of compatibility between the first content item and the second candidate metaverse environment; and determining that the first compatibility score is higher than the second compatibility score; wherein the first candidate metaverse environment is selected as the target metaverse environment for placement of the first content item based at least in part on determining that the first compatibility score is higher than the second compatibility score (Goyal: FIG. 5:520 and ¶[0038]…Goyal computes a separate compatibility (confidence) score for each of plural candidates and selects the candidate having the highest score, and Hyndman likewise evaluates plural predefined candidate spaces of the environment, so applied to plural candidate environments in the combination this constitutes computing a second compatibility score, determining the first exceeds the second, and selecting the higher-scoring environment for placement, "Machine learning model 520 calculates a confidence for each input candidate object. The candidate object having the highest confidence is then selected and transmitted 532 to virtual environment generation circuitry 506, which incorporates the selected candidate object into the virtual environment"). The rationale to combine Hyndman with Goyal is the same as the parent claim.
Per claim 3, Goyal combined with Hyndman discloses claim 1. Goyal further teaches wherein applying the machine learning model comprises: computing an environment feature vector based on keywords associated with the first candidate metaverse environment (Goyal: FIG. 4 and ¶[0027]…Goyal applies a word-embedding model to the attributes of the environment's virtual objects to produce input vectors for the model, which constitutes computing an environment feature vector based on keywords associated with the environment under BRI, "In some embodiments, a word embedding model, such as Word2Vec, is applied to the attributes to generate corresponding vectors for input into the machine learning model").
Per claim 4, Goyal combined with Hyndman discloses claim 3. Hyndman further teaches determining the keywords associated with the first candidate metaverse environment by scraping information from the first candidate metaverse environment (Hyndman: Fig. 2:S34 and ¶[0031]…Hyndman's server extracts context metadata directly from within the three-dimensional environment, which constitutes scraping information from the environment to determine its keywords under BRI, "Virtual environment server 16 extracts any 3D context data, i.e., metadata, within the 3D environment, at step S34"). The rationale to combine Hyndman with Goyal is the same as the parent claim.
Per claim 5, Goyal combined with Hyndman discloses claim 4. Hyndman further teaches identifying the keywords associated with the first candidate metaverse environment based on or more of: metadata associated with the first candidate metaverse environment; code associated with the first candidate metaverse environment; subject matter displayed in the first candidate metaverse environment; and objects included in the first candidate metaverse environment (Hyndman: Fig. 2:S34 and ¶[0032]…Hyndman derives the keywords from environment metadata (room names), object names, resource file names (code), and content displayed in the environment, which constitutes each of the recited alternative keyword sources under BRI, "The metadata can include volume names, e.g., room names, texture and object names such as labels and resource file names, and document file name and content, e.g., from a PowerPoint document displayed in the environment"). The rationale to combine Hyndman with Goyal is the same as the parent claim.
Per claim 6, Goyal combined with Hyndman discloses claim 1. Goyal further teaches wherein applying the machine learning model comprises: computing a content feature vector based on keywords associated with the first content item (Goyal: FIG. 4 and ¶[0027]…Goyal applies the same word-embedding model to the attributes of the candidate (content) object to produce its input vector, which constitutes computing a content feature vector based on keywords associated with the content item under BRI, "In some embodiments, a word embedding model, such as Word2Vec, is applied to the attributes to generate corresponding vectors for input into the machine learning model").
Per claim 7, Goyal combined with Hyndman discloses claim 6. Goyal further teaches determining the keywords associated with the first content item by scraping information from the first content item (Goyal: FIG. 5:516 and ¶[0029]…Goyal extracts the object attributes of each candidate content object directly from the virtual environment data by object identification, which constitutes determining the content item's keywords by scraping information from the content item under BRI, "Virtual object identification circuitry 516 may extract from the virtual environment data metadata describing each virtual object, which may include object attributes such as those described above in connection with FIG. 3").
Claims 8, 9, 10, 11, 12, 13 and 14 are substantially similar in scope and spirit as claims 1, 2, 3, 4, 5, 6 and 7, respectively. Therefore the rejections of claims 1, 2, 3, 4, 5, 6 and 7 are applied accordingly.
Claims 15, 16, 17, 18 and 19 are substantially similar in scope and spirit as claims 1, 2, 3, 4 and 5, respectively, and claim 20 is substantially similar in scope and spirit as claims 6 and 7. Therefore the rejections of claims 1, 2, 3, 4, 5, 6 and 7 are applied accordingly.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ALAN CHEN whose telephone number is (571)272-4143. The examiner can normally be reached M-F 10-7.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Kamran Afshar can be reached at (571) 272-7796. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/ALAN CHEN/Primary Examiner, Art Unit 2125
Prosecution Insights