DETAILED ACTION
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 .
This action is in response to the application and claims filed 5/24/2023. Claims 1-20 are pending and have been examined. Claims 1-20 are rejected.
Priority
Applicant’s claim for the benefit of a prior-filed application under 35 U.S.C. 119(e) or under 35 U.S.C. 120, 121, 365(c), or 386(c) is acknowledged. The present application is a continuation application of PCT application number PCT/KR22/13026, filed on 8/31/2022, which claims foreign priority to Korean patent application No. KR10-2022-0045188 filed on 4/12/2022 and Korean patent application No. KR10-2021-0115908 filed on 8/31/2021.
The examiner acknowledges that certified copies of Korean patent application Nos. KR10-2022-0045188 and KR10-2021-0115908 have been retrieved (on 7/03/2023, in Korean). The examiner notes that translations of Korean patent application Nos. KR10-2022-0045188 and KR10-2021-0115908 do not appear to have been furnished to-date.
Information Disclosure Statement
Acknowledgment is made of the information disclosure statements filed 5/24/2023, 11/20/2023 and 12/18/2024, which comply with 37 CFR 1.97. As such, the information disclosure statements have been placed in the application file and the information referred to therein has been considered by the examiner.
Drawings
The drawings are also objected to as failing to comply with 37 CFR 1.84(p)(3) because Figures 2, 5 and 7 include letters which do not measure at least .32 cm. (1/8 inch) in height (i.e., most of the lowercase, subscript and superscript characters in FIGs. 2, 5 and 7).
The drawings are also objected to as failing to comply with 37 CFR 1.84(p)(5) because they include the following reference characters not mentioned in the description:
130, 140 and 150 in FIG. 4 (see, paragraphs 119-130 describing FIG. 4, which includes reference signs 100, 110 and 120 but not 130, 140 and 150).
Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. The figure or figure number of an amended drawing should not be labeled as “amended.” If a drawing figure is to be canceled, the appropriate figure must be removed from the replacement sheet, and where necessary, the remaining figures must be renumbered and appropriate changes made to the brief description of the several views of the drawings for consistency. Additional replacement sheets may be necessary to show the renumbering of the remaining figures. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Specification
The disclosure is objected to because of the following informalities:
Reference characters 130, 140 and 150 shown in Figure 4 are not found in the detailed description (see, e.g., paragraphs 119-130 describing FIG. 4). Appropriate correction is required.
The title of the invention is objected to as not being descriptive. A new title is required that is clearly indicative of the invention to which the claims are directed. In particular, the title of the invention is “ELECTRONIC DEVICE AND CONTROLLING METHOD OF ELECTRONIC DEVICE”; however the title is not descriptive of the claimed invention. As such, the examiner believes that the title of the invention is imprecise. A descriptive title indicative of the invention will help in proper indexing, classifying, searching, etc. See, MPEP § 606.01. However, the title of the invention should be limited to 500 characters. The examiner suggests including the aspect(s) of the claims which Applicant believes to be novel or nonobvious over the prior art.
Claim Objections
Claims 17-20 are objected to because of the following informalities:
Lines 11-12 of independent claim 17 recite “obtaining a plurality of augmented vectors based on the plurality of augmented graphs”. For consistency with other operations in the claim, the recitation of gerund “obtaining” should read “obtain” (see, e.g., the 1st and 2nd operations of the claim and line 10 of the claim reciting “obtain”). Appropriate correction is required.
Line 10 of claim 17 recites “obtain a plurality of augmented graphs based on each of the integrated graph”. This recitation is grammatically incorrect. If supported by Applicant’s original specification, the examiner suggests that one way to address this objection would be to amend “obtain a plurality of augmented graphs based on each of the integrated graph” to read “obtain a plurality of augmented graphs based on each of the integrated graphs”. Appropriate correction is required.
Claims 18-20, which each depend directly or indirectly from claim 17, are objected to based on their respective dependencies from claim 17.
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.
Claims 6 and 14 are rejected under 35 U.S.C. 112(b) as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention.
Claims 6 and 14 both recite “vectors in a short distance on a vector space, and map augmented graphs corresponding to different individual graphs among the plurality of augmented graphs with vectors in a long distance on the vector space.” (see, lines 3-6 of claims 6 and 14). The terms “a short distance on a vector space” and “a long distance on the vector space” in dependent claims 6 and 14 are relative terms which renders the claims indefinite. The terms “a short distance on a vector space" and “a long distance on the vector space” are not defined by the claims, the specification does not provide a standard for ascertaining the requisite degree, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. In particular, it is unclear what metrics are used for ascertaining the requisite degree or distance measurements are covered by the terms “a short distance” and “a long distance” in the phrases “a long distance on the vector space.” The specification merely repeats the claim language in paragraphs 12 and 20 and mentions a general examples in paragraphs 93 and 102 in stating “applying different augmentation methods to data and training the neural network model 10 so that a distance between feature representations of a positive pair among the data is short and a distance between feature representations of a negative pair among the data is long.” and “Referring to FIG. 2, for example, the neural network model 10 may be trained to map z1a1 with z1a2 and z1a3 in a relationship of the positive pair as vectors in a short distance, and to map z1 al with z2a1, Z3a1, and Z4a1 in a relationship of the negative pair as vectors in a long distance.” However, the specification does not explicitly define what is meant by the recited “a large portion of the embedding space” or provide a standard for ascertaining the requisite degree of the relative terms “short” or “long” of the claimed “a short distance on a vector space” and “a long distance on the vector space.” Applicant’s specification also does not provide a standard for ascertaining the requisite distance measurements or amounts of “a short distance on a vector space" and “a long distance on the vector space” recited in claims 6 and 14. For examination purposes, the claimed “short distance on a vector space" is being interpreted as any distance measurement or amount that is less than the recited “long distance on the vector space”. Appropriate correction is required.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f):
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f). The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f), is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f). The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f), is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f), except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f), except as otherwise indicated in an Office action.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f), because the claim limitations use a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitations are:
a prediction module configured to obtain recommended content information,
an augmentation module configured to obtain the plurality of augmented graphs, and a contrastive loss obtaining module configured to obtain the contrastive loss in claim 5.
Regarding claim 5 and the above-noted three-prong test, the recited prediction module is a generic placeholder, configured to obtain recommended content information is functional language, and there is no recitation in the claim of sufficient structure to perform the obtaining. With regard to claim 5 and the above-noted three-prong test, the recited augmentation module is a generic placeholder, configured to obtain the plurality of augmented graphs is functional language, and there is no recitation in the claim of sufficient structure to perform the obtaining. Further regarding claim 5, and the above-noted three-prong test, the recited contrastive loss obtaining module is a generic placeholder, configured to obtain the contrastive loss is functional language, and there is no recitation in the claim of sufficient structure to perform the obtaining.
Because these claim limitations are being interpreted under 35 U.S.C. 112(f) they are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
A review of the specification shows that the following appears to be the corresponding structure described in the specification for the 35 U.S.C. 112(f) limitations:
Regarding the modules recited in claim 5, with reference to the modules depicted in FIG. 2, paragraphs 49, 65, 75-78, 84, 88, 90, 92, 94-97, 100-101, 103 and 116-117 describe the corresponding structure of modules capable of performing the claimed functions.
If applicant does not intend to have these limitations interpreted under 35 U.S.C. 112(f), applicant may: (1) amend the claim limitations to avoid them being interpreted under 35 U.S.C. 112(f) (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitations recite sufficient structure to perform the claimed function so as to avoid them being interpreted under 35 U.S.C. 112(f).
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. because the claimed invention is directed to an abstract idea without significantly more. The analysis below of the claims’ subject matter eligibility follows the 2019 Revised Patent Subject Matter Eligibility Guidance, 84 Fed. Reg. 50-57 (January 7, 2019) (“2019 PEG”) and the 2024 Guidance Update on Patent Subject Matter Eligibility, Including on Artificial Intelligence, 89 Fed. Reg. 58128-58138 (July 17, 2024) (“2024 AI SME Update”).
When considering subject matter eligibility under 35 U.S.C. 101, it must be determined whether the claim is directed to one of the four statutory categories of invention, i.e., process, machine, manufacture, or composition of matter (Step 1). If the claim does fall within one of the statutory categories, the second step in the analysis is to determine whether the claim is directed to a judicial exception (Step 2A). The Step 2A analysis is broken into two prongs. In the first prong (Step 2A, Prong 1), it is determined whether or not the claims recite a judicial exception (e.g., mathematical concepts, mental processes, certain methods of organizing human activity). If it is determined in Step 2A, Prong 1 that the claims recite a judicial exception, the analysis proceeds to the second prong (Step 2A, Prong 2), where it is determined whether or not the claims integrate the judicial exception into a practical application. If it is determined at step 2A, Prong 2 that the claims do not integrate the judicial exception into a practical application, the analysis proceeds to determining whether the claim is a patent-eligible application of the exception (Step 2B). If an abstract idea is present in the claim, any element or combination of elements in the claim must be sufficient to ensure that the claim integrates the judicial exception into a practical application, or else amounts to significantly more than the abstract idea itself.
Regarding independent claims 1, 9 and 16, these claims are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 1 is directed to an electronic device comprising a memory and a processor, corresponding to an article of manufacture, claim 9 is directed to a method, corresponding to a process, and claim 16 is directed to a non-transitory computer readable storage medium, corresponding to an article of manufacture, which are each one of the statutory categories.
Step 2A Prong One Analysis: The claims are directed to an abstract idea. In particular, the claims recite mental processes that are concepts performed in the human mind (including an observation, evaluation, judgment, opinion).
The limitations recited in claims 1, 9 and 16, using respective similar language:
generate an integrated graph, in which the plurality of individual graphs are integrated, based on a connection relationship between nodes included in the plurality of individual graphs and a number of times each connection between the nodes is repeated;
provide recommended content based on the plurality of augmented graphs - as drafted, under their broadest reasonable interpretation (BRI), in view of the specification, cover concepts performed in the human mind (evaluation, judgement, or opinion to generate/create an integrated graph based on observed relationships and connections between data/nodes in an obtained set of individual graphs, and to recommend content based on observed data in augmented graphs).
The above limitations in the context of these claims encompass, inter alia, generating/creating an integrated graph based on observed relationships and connections between data/nodes in a received/obtained set of graphs, and to recommend content based on observed data in augmented graphs (evaluation/judgement/opinion to link/correlate and relate nodes in observed/received graphs to create an integrated graph; and then using an augmented graph to recommend content) (corresponding to mental processes which can be done mentally or by pen and paper).
Regarding the “integrated graph”, “individual graphs” and “augmented graphs”, aside from including nodes and data “representing an access history of a user for a plurality of contents” and having “a connection relationship between nodes included in the plurality of individual graphs”, no details of the graphs are recited and the graphs are recited at a high level of generality and can be constructed by hand with pen and paper. Thus, the claimed “integrated graph”, “individual graphs” and “augmented graphs”, under the BRI, in light of the specification, could be constructed by hand with pen and paper and then manually modified/augmented by hand with pen and paper based on a reasonable amount of observed data (i.e., “an access history of a user for a plurality of contents” and “connection relationship between nodes included in the plurality of individual graphs and a number of times each connection between the nodes is repeated”).
Step 2A Prong Two Analysis: The judicial exceptions are not integrated into a practical application.
In particular, these claims recite, using respective similar language, the additional elements of a memory storing a neural network model (claim 1),
storage medium storing program instructions (claim 16),
obtain a plurality of individual graphs representing an access history of a user for a plurality of contents by a plurality of sessions; and
obtain a plurality of augmented graphs by augmenting each of the plurality of individual graphs based on the integrated graph – These are insignificant extra-solution activities that are not integrated into the claims as a whole and do not add a meaningful limitation to the above-noted mental processes specified in these claims. That is, “storing a neural network model”, “storing program instructions”, “obtain a plurality of individual graphs representing an access history” and “obtain a plurality of augmented graphs by augmenting each of the plurality of individual graphs” is mere data gathering and storage (See MPEP § 2106.05(g)).
The claims also recite, using respective similar language, the additional elements: An electronic device comprising: a memory storing a neural network model; and a processor configured to: <perform operations> (claim 1), A non-transitory computer readable storage medium storing program instructions that are executable by a processor, wherein the program instructions, when executed by the processor, are configured to control a device to: <perform operations> (claim 16), and train the neural network model to provide recommended content - The above-noted additional elements in the claims amount to recitation of the words "apply it" (or an equivalent) or are mere instructions to implement an abstract idea or other exception on a computer, which does not integrate a judicial exception into a practical application. See MPEP 2106.05(f). In particular, “train the neural network model to provide recommended content” is simply generic training to perform the abstract idea of processing data to provide recommended content and amounts to mere instructions to apply the exception (MPEP 2106.05(f)).
Merely asserting that a judicial exception is to be carried out on a generic computer (i.e., with the generically-recited “neural network” and the “device” and “processor” of claims 1 and 16) cannot meaningfully integrate the judicial exception into a practical application. See MPEP § 2106.05(f).
Step 2B Analysis: The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above with respect to the integration of the abstract idea into a practical application, the additional elements of:
“An electronic device comprising: a memory storing a neural network model; and a processor configured to:” <perform operations> (claim 1) and “A non-transitory computer readable storage medium storing program instructions that are executable by a processor, wherein the program instructions, when executed by the processor, are configured to control a device to:” <perform operations> (claim 16) amount to no more than mere instructions to apply the exception (MPEP 2106.05(f))
generic training to perform the abstract idea amounts to no more than mere instructions to apply the exception (MPEP 2106.05(f))
Mere instructions to apply the mental process electronically (i.e., with the generically-recited “neural network” and the “device” and “processor” of claims 1 and 16) do not amount to significantly more than the judicial exception. As noted above, merely asserting that a judicial exception is to be carried out on a generic computer cannot provide significantly more than the judicial exception. See MPEP § 2106.05(f).
Moreover, receiving, communicating, forwarding and storing data are insignificant extra-solution activities that are well-understood, routine, and conventional. See MPEP2106.05(d)(II) (“The courts have recognized the following computer functions as well-understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity. i. Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); … iv. Storing and retrieving information in memory”) (citing OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015)) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network)". Therefore, the recitations of “a memory storing a neural network model” (claim 1), “storing program instructions” (claim 16), “obtain[ing] a plurality of individual graphs representing an access history of a user for a plurality of contents by a plurality of sessions” and “obtain[ing] a plurality of augmented graphs by augmenting each of the plurality of individual graphs based on the integrated graph” are the well-understood, routine, conventional activities of storing information in memory, and receiving and transmitting data over a network, as discussed in MPEP § 2106.05(d).
Accordingly, at Step 2B, the additional elements do not amount to significantly more than the judicial exception.
As an ordered whole, the claims are directed to a method of generating/creating an integrated graph based on observed relationships and connections between data/nodes in an obtained set of individual graphs, and then recommending content based on observed data in augmented graphs. Nothing in the claims provide significantly more than this.
The additional elements do not provide an inventive concept, and, therefore, the claims are not patent eligible.
Regarding independent claim 17, this claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 17 is directed to a device comprising a memory and a processor, corresponding to an article of manufacture, which is one of the statutory categories.
Step 2A Prong One Analysis: The claim is directed to an abstract idea. In particular, the claim recites mental processes that are concepts performed in the human mind (including an observation, evaluation, judgment, opinion).
The limitation recited in claim 17:
identify recommended content by controlling a neural network model based on the plurality of vectors; - as drafted, under its broadest reasonable interpretation (BRI), in view of the specification, covers concepts performed in the human mind (evaluation, judgement, or opinion to identify recommended content using a generically-recited neural network model based on observed vector data).
The above limitation in the context of this claim encompasses recommending content based on observed data in vectors (evaluation/judgement/opinion to recommend content based on data in vectors) (corresponding to mental processes which can be done mentally or by pen and paper).
Regarding the “neural network model”, no details of the model are recited and the model is recited at a high level of generality and can be constructed by hand with pen and paper. Thus, the claimed “neural network model” under the BRI, in light of the specification, could be constructed by hand with pen and paper and then manually controlled by hand with pen and paper based on a reasonable amount of observed data (i.e., “access history corresponding to a plurality of contents” and data in “the plurality of vectors”).
Step 2A Prong Two Analysis: The judicial exceptions are not integrated into a practical application.
In particular, this claim recites the additional elements of
instructions stored in the memory;
obtain a plurality of individual graphs based on access history corresponding to a plurality of contents;
obtain a plurality of vectors based on the plurality of individual graphs;
obtain an integrated graph based on each of the plurality of individual graphs;
obtain a plurality of augmented graphs based on each of the integrated graph;
obtaining a plurality of augmented vectors based on the plurality of augmented graphs – These are insignificant extra-solution activities that are not integrated into the claim as a whole and do not add a meaningful limitation to the above-noted mental processes specified in this claim. That is the above-noted “instructions stored in the memory”, “obtain” and “obtaining” operations are mere data gathering and storage (See MPEP § 2106.05(g)).
Regarding the “integrated graph”, “individual graphs” and “augmented graphs”, aside from including “access history corresponding to a plurality of contents”, no details of the graphs are recited and the graphs are recited at a high level of generality and can be constructed by hand with pen and paper. Thus, the claimed “integrated graph”, “individual graphs” and “augmented graphs”, under the BRI, in light of the specification, could be constructed by hand with pen and paper and then manually modified/augmented by hand with pen and paper based on a reasonable amount of observed data (i.e., “access history corresponding to a plurality of contents”).
The claim also recites the additional elements: A device comprising: a memory storing a neural network model; and a processor configured to execute instructions stored in the memory to: <perform operations> and train the neural network model based on the plurality of augmented vectors - The above-noted additional elements in the claim amount to recitation of the words "apply it" (or an equivalent) or are mere instructions to implement an abstract idea or other exception on a computer, which does not integrate a judicial exception into a practical application. See MPEP 2106.05(f). In particular, “train the neural network model based on the plurality of augmented vectors” is simply generic training to perform the abstract idea of processing data in the observed, augmented vectors and amounts to mere instructions to apply the exception (MPEP 2106.05(f)).
Merely asserting that a judicial exception is to be carried out on a generic computer (i.e., with the generically-recited “neural network” and “device comprising: a memory; and a processor”) cannot meaningfully integrate the judicial exception into a practical application. See MPEP § 2106.05(f).
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above with respect to the integration of the abstract idea into a practical application, the additional elements of:
“A device comprising: a memory storing a neural network model; and a processor configured to execute instructions stored in the memory to: <perform operations> amounts to no more than mere instructions to apply the exception (MPEP 2106.05(f))
generic training to perform the abstract idea amounts to no more than mere instructions to apply the exception (MPEP 2106.05(f))
The additional elements do not provide an inventive concept, and, therefore, the claim is not patent eligible.
Mere instructions to apply the mental process electronically (i.e., with the generically-recited “neural network” and the generic “device comprising: a memory; and a processor”) do not amount to significantly more than the judicial exception. As noted above, merely asserting that a judicial exception is to be carried out on a generic computer cannot provide significantly more than the judicial exception. See MPEP § 2106.05(f).
Moreover, receiving, communicating, forwarding and storing data are insignificant extra-solution activities that are well-understood, routine, and conventional. See MPEP2106.05(d)(II) (“The courts have recognized the following computer functions as well-understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity. i. Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); … iv. Storing and retrieving information in memory”) (citing OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015)) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network)". Therefore, the recitations of “instructions stored in the memory”, “obtain a plurality of individual graphs based on access history corresponding to a plurality of contents; obtain a plurality of vectors based on the plurality of individual graphs; obtain an integrated graph based on each of the plurality of individual graphs; obtain a plurality of augmented graphs based on each of the integrated graph” and “obtaining a plurality of augmented vectors based on the plurality of augmented graphs” are the well-understood, routine, conventional activities of storing information in memory, and receiving and transmitting data over a network, as discussed in MPEP § 2106.05(d).
Accordingly, at Step 2B, the additional elements do not amount to significantly more than the judicial exception.
As an ordered whole, the claim is directed to a method of generating/creating an integrated graph based on observed relationships and connections between data/nodes in an obtained set of individual graphs, and then recommending content based on observed data in augmented graphs. Nothing in the claim provides significantly more than this. As such, the claim is not patent eligible.
Regarding claim 18, this claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 18 is directed to a device as depending from claim 17, thus the analysis for patent eligibility of claim 17 is incorporated herein.
Step 2A Prong 1: The claim recites obtain a main loss by comparing the recommended content identified based on the plurality of vectors with label information - This limitation, as drafted, under their BRI, in view of the specification, covers concepts performed in the human mind (evaluation, judgement, or opinion to obtain/identify a main loss by comparing recommended content with label information, where the recommended content is based on observed vector data).
Step 2A Prong Two Analysis: The judicial exceptions are not integrated into a practical application.
In particular, this claim recites the additional element of wherein the processor is further configured to execute the instructions stored in the memory to obtain a main loss – This additional element amounts to recitation of the words "apply it" (or an equivalent) or are mere instructions to implement an abstract idea or other exception on a computer, which does not integrate a judicial exception into a practical application. See MPEP 2106.05(f).
Merely asserting that a judicial exception is to be carried out on a generic computer (i.e., with the generically-recited “processor” and “memory”) cannot meaningfully integrate the judicial exception into a practical application. See MPEP § 2106.05(f).
The claim also recites the additional element: label information stored in the memory – This is an insignificant extra-solution activity that is not integrated into the claim as a whole and does not add a meaningful limitation to the above-noted mental process specified in this claim. That is the above-noted “label information stored in the memory” is mere data storage (See MPEP § 2106.05(g)).
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above with respect to the integration of the abstract idea into a practical application, mere instructions to apply the mental process electronically (i.e., with the generically-recited “processor” and “memory”) do not amount to significantly more than the judicial exception. As noted above, merely asserting that a judicial exception is to be carried out on a generic computer cannot provide significantly more than the judicial exception. See MPEP § 2106.05(f).
Moreover, receiving, communicating, forwarding and storing data are insignificant extra-solution activities that are well-understood, routine, and conventional. See MPEP2106.05(d)(II) (“The courts have recognized the following computer functions as well-understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity. i. Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); … iv. Storing and retrieving information in memory”) (citing OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015)) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network)". Therefore, the recitation of “label information stored in the memory” is the well-understood, routine, conventional activity of storing information in memory, as discussed in MPEP § 2106.05(d).
Accordingly, at Step 2B, the additional elements do not amount to significantly more than the judicial exception.
As an ordered whole, the claim is directed to a method of obtaining/identifying a main loss by comparing recommended content with label information, where the recommended content is based on observed vector data. Nothing in the claim provides significantly more than this. As such, the claim is not patent eligible.
Regarding claim 19, this claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 19 is directed to a device as depending from claim 18, thus the analysis for patent eligibilities of claim 18 and base claim 17 are incorporated herein.
Step 2A Prong 1: See claims 17 and 18 above.
Step 2A Prong Two Analysis: The judicial exceptions are not integrated into a practical application.
The claim recites wherein the processor is further configured to execute the instructions stored in the memory to train the neural network model based on the main loss - This additional element amounts to recitation of the words "apply it" (or an equivalent) or are mere instructions to implement an abstract idea or other exception on a computer, which does not integrate a judicial exception into a practical application. See MPEP 2106.05(f).
In particular, this claim recites the additional element of wherein the processor is further configured to execute the instructions … to obtain a main loss – This additional element amounts to recitation of the words "apply it" (or an equivalent) or are mere instructions to implement an abstract idea or other exception on a computer, which does not integrate a judicial exception into a practical application. See MPEP 2106.05(f).
Merely asserting that a judicial exception is to be carried out on a generic computer (i.e., with the generically-recited “processor” and “memory”) cannot meaningfully integrate the judicial exception into a practical application. See MPEP § 2106.05(f).
The claim also recites the additional element: instructions stored in the memory – This is an insignificant extra-solution activity that is not integrated into the claim as a whole and does not add a meaningful limitation to the above-noted mental process specified in this claim. That is the above-noted “instructions stored in the memory” is mere data storage (See MPEP § 2106.05(g)).
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above with respect to the integration of the abstract idea into a practical application, mere instructions to apply the mental process electronically (i.e., with the generically-recited “processor” and “memory”) do not amount to significantly more than the judicial exception. As noted above, merely asserting that a judicial exception is to be carried out on a generic computer cannot provide significantly more than the judicial exception. See MPEP § 2106.05(f).
Moreover, receiving, communicating, forwarding and storing data are insignificant extra-solution activities that are well-understood, routine, and conventional. See MPEP2106.05(d)(II) (“The courts have recognized the following computer functions as well-understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity. i. Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); … iv. Storing and retrieving information in memory”) (citing OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015)) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network)". Therefore, the recitation of “instructions stored in the memory” is the well-understood, routine, conventional activity of storing information in memory, as discussed in MPEP § 2106.05(d).
Accordingly, at Step 2B, the additional elements do not amount to significantly more than the judicial exception.
Thus, the claim is not patent eligible.
Regarding claims 2, 10 and 20, these claims are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claims 2 and 20 are directed to devices as depending from claims 1 and 17, respectively, and claim 10 is directed to a method as depending from claim 9, thus the analysis for patent eligibilities of claims 1, 9 and 17 are incorporated herein.
Step 2A Prong 1: See independent claims 1, 9 and 17 above.
Step 2A Prong Two Analysis: The judicial exceptions are not integrated into a practical application.
The claims recite, using respective similar language, the additional elements: obtain at least one augmented graph of the plurality of augmented graphs by adding at least one node included in the integrated graph to a plurality of nodes included in each of the plurality of individual graphs - this is an insignificant extra-solution activity that in not integrated into the claims as a whole and does not add a meaningful limitation to the above-noted mental processes specified in these claims. That is, “obtain[ing] at least one augmented graph of the plurality of augmented graphs by adding at least one node” is mere data gathering (See MPEP § 2106.05(g)).
Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claims are directed to an abstract idea.
Step 2B Analysis: The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Receiving, communicating and forwarding data are insignificant extra-solution activities that are well-understood, routine, and conventional. See MPEP2106.05(d)(II) (“The courts have recognized the following computer functions as well-understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity. i. Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); … iv. Storing and retrieving information in memory”) (citing OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015)) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network)". Therefore, recitations of “obtain[ing] at least one augmented graph of the plurality of augmented graphs by adding at least one node included in the integrated graph to a plurality of nodes included in each of the plurality of individual graphs” are the well-understood, routine, conventional activities of receiving and transmitting data over a network as discussed in MPEP § 2106.05(d).
Accordingly, at Step 2B, the additional elements do not amount to significantly more than the judicial exception.
Regarding claims 3 and 11, these claims are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 3 is directed to a device as depending from claim 1, and claim 11 is directed to a method as depending from claim 9, thus the analysis for patent eligibilities of claims 1 and 9 are incorporated herein.
Step 2A Prong 1: The claims recite, using respective similar language, obtain at least one augmented graph of the plurality of augmented graphs by changing at least one node of a plurality of nodes included in each of the plurality of individual graphs to at least one node included in the integrated graph - These limitations, as drafted, under their BRI, in view of the specification, cover concepts performed in the human mind (evaluation, judgement, or opinion to obtain an augmented graph by changing/modifying a node in each of the individual graphs to a node in the integrated graph).
The above limitations in the context of these claims encompass, inter alia, obtaining an augmented graph by reassigning/changing one or more nodes in the individual graphs to be nodes in the integrated graph (corresponding to mental processes which can be done mentally or by pen and paper).
Regarding “the augmented graphs” and “individual graphs”, no details of the graphs are recited and the graphs are recited at a high level of generality, and can be constructed by hand with pen and paper. Thus, the claimed graphs, under the BRI, in light of the specification, could be constructed by hand with pen and paper and then manually modified/updated by hand with pen and paper based on a reasonable amount of observed data (i.e., the nodes).
Step 2A Prong Two Analysis: The judicial exceptions are not integrated into a practical application.
The claims do not recite any additional limitations or elements which integrate the abstract idea into a practical application.
There are no additional elements to integrate the abstract idea into a practical application or to impose any meaningful limits on practicing the abstract idea. The claims are directed to an abstract idea.
Step 2B Analysis: The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. Viewing the additional elements of these dependent claims as a combination does not add anything further than the individual elements.
This claims are not patent eligible.
Regarding claims 4 and 12, these claims are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 4 is directed to a device as depending from claim 1, and claim 12 is directed to a method as depending from claim 9, thus the analysis for patent eligibilities of claims 1 and 9 are incorporated herein.
Step 2A Prong 1: See independent claims 1 and 9 above.
Step 2A Prong Two Analysis: The judicial exceptions are not integrated into a practical application.
In particular, these claims recite, using respective similar language, the additional element of train the neural network model based on a contrastive loss for the plurality of augmented graphs.
The above-noted additional elements in the claims amount to recitation of the words "apply it" (or an equivalent) or are mere instructions to implement an abstract idea or other exception on a computer, which does not integrate a judicial exception into a practical application. See MPEP 2106.05(f). In particular, “train the neural network model based on a contrastive loss for the plurality of augmented graphs” is simply generic training to perform the abstract idea of processing and contrasting data (i.e., evaluation/judgment/option to compare/contrast data in the augmented graphs) and amounts to mere instructions to apply the exception (MPEP 2106.05(f)).
Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claims are directed to an abstract idea.
Merely asserting that a judicial exception is to be carried out on a generic computer (i.e., with the generically-recited “neural network” and the “device comprising a memory … and a processor” of base claim 1) cannot meaningfully integrate the judicial exception into a practical application. See MPEP § 2106.05(f).
Step 2B Analysis: The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above with respect to the integration of the abstract idea into a practical application, the additional elements of:
generic training to perform the abstract idea amounts to no more than mere instructions to apply the exception (MPEP 2106.05(f))
Mere instructions to apply the mental process electronically (i.e., with the generically-recited “neural network” and the “device comprising a memory … and a processor” of base claim 1) do not amount to significantly more than the judicial exception. As noted above, merely asserting that a judicial exception is to be carried out on a generic computer cannot provide significantly more than the judicial exception. See MPEP § 2106.05(f).
Accordingly, at Step 2B, the additional element does not amount to significantly more than the judicial exception.
The additional element does not provide an inventive concept, and, therefore, the claims are not patent eligible.
Regarding claims 5 and 13, these claims are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 5 is directed to a device as depending from claim 4, and claim 13 is directed to a method as depending from claim 11, thus the analysis for patent eligibilities of claims 4 and 12, and of base claims 1 and 9 are incorporated herein.
Step 2A Prong 1: The claims recite, using respective similar language,
obtain recommended content information corresponding to each of the plurality of individual graphs, and
obtain the contrastive loss for the plurality of augmented graphs – as drafted, under their BRI, in view of the specification, cover concepts performed in the human mind (evaluation, judgement, or opinion to obtain recommended content/recommend content based on observed data in individual graphs, and to obtain/identify contrastive loss based on observed data in the augmented graphs).
The above limitations in the context of these claims encompass, inter alia, recommending content based on observed data in individual graphs and comparing/contrasting observed data in the augmented graphs (evaluation/judgement/opinion to use individual graphs to recommend content and then compare/contrast observed data in the augmented graphs) (corresponding to mental processes which can be done mentally or by pen and paper).
Regarding the “individual graphs” and “augmented graphs”, aside from including nodes, data and “content information”, no details of the graphs are recited and the graphs are recited at a high level of generality and can be constructed by hand with pen and paper. Thus, the claimed “individual graphs” and “augmented graphs”, under the BRI, in light of the specification, could be constructed by hand with pen and paper based on a reasonable amount of observed data (i.e., “content information”).
Step 2A Prong Two Analysis: The judicial exceptions are not integrated into a practical application.
In particular, these claims recite, using respective similar language, the additional elements of obtain vectors corresponding to the plurality of individual graphs and vectors corresponding to the plurality of augmented graphs,
obtain recommended content information,
obtain the plurality of augmented graphs by augmenting each of the plurality of individual graphs, and
obtain the contrastive loss – These are insignificant extra-solution activities that are not integrated into the claims as a whole and do not add a meaningful limitation to the above-noted mental processes specified in these claims. That is, the above-noted “obtain vectors corresponding to the plurality of individual graphs and vectors corresponding to the plurality of augmented graphs”, “obtain recommended content information”, “obtain the plurality of augmented graphs by augmenting each of the plurality of individual graphs” and “obtain the contrastive loss” is mere data gathering (See MPEP § 2106.05(g)).
The claims also recite, using respective similar language, the additional elements: implement an encoder configured to obtain vectors, a prediction module configured to obtain recommended content information, an augmentation module configured to obtain the plurality of augmented graphs … and a contrastive loss obtaining module configured to obtain the contrastive loss - These additional elements in the claims amount to recitation of the words "apply it" (or an equivalent) or are mere instructions to implement an abstract idea or other exception on a computer, which does not integrate a judicial exception into a practical application. See MPEP 2106.05(f). In particular, “train the neural network model to provide recommended content” is simply generic training to perform the abstract idea of processing data to provide recommended content and amounts to mere instructions to apply the exception (MPEP 2106.05(f)).
Merely asserting that a judicial exception is to be carried out on a generic computer (i.e., with the generically-recited “encoder”, modules and the “device comprising: a memory … and a processor” of claim 1) cannot meaningfully integrate the judicial exception into a practical application. See MPEP § 2106.05(f).
Step 2B Analysis: The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above with respect to the integration of the abstract idea into a practical application, the additional elements of:
“implement an encoder configured to obtain vectors, a prediction module configured to obtain recommended content information, an augmentation module configured to obtain the plurality of augmented graphs … and a contrastive loss obtaining module configured to obtain the contrastive loss” amount to no more than mere instructions to apply the exception (MPEP 2106.05(f)).
Mere instructions to apply the mental process electronically (i.e., with the generically-recited “encoder”, modules and the “device comprising: a memory … and a processor” of base claim 1) do not amount to significantly more than the judicial exception. As noted above, merely asserting that a judicial exception is to be carried out on a generic computer cannot provide significantly more than the judicial exception. See MPEP § 2106.05(f).
Moreover, receiving, communicating, forwarding and storing data are insignificant extra-solution activities that are well-understood, routine, and conventional. See MPEP2106.05(d)(II) (“The courts have recognized the following computer functions as well-understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity. i. Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); … iv. Storing and retrieving information in memory”) (citing OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015)) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network)". Therefore, the recitations of “obtain vectors corresponding to the plurality of individual graphs and vectors corresponding to the plurality of augmented graphs”, “obtain recommended content information”, “obtain the plurality of augmented graphs by augmenting each of the plurality of individual graphs” and “obtain the contrastive loss” are the well-understood, routine, conventional activities of receiving and transmitting data over a network, as discussed in MPEP § 2106.05(d).
Accordingly, at Step 2B, the additional elements do not amount to significantly more than the judicial exception.
Regarding claims 6 and 14, these claims are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 6 is directed to a device as depending from claim 5, and claim 14 is directed to a method as depending from claim 13, thus the analysis for patent eligibilities of claims 5 and 13, of intervening claims 4 and 12, and of base claims 1 and 9 are incorporated herein.
Step 2A Prong 1: The claims recite, using respective similar language,
map augmented graphs corresponding to common individual graphs, among the plurality of augmented graphs, with vectors in a short distance on a vector space, and map augmented graphs corresponding to different individual graphs among the plurality of augmented graphs with vectors in a long distance on the vector space1 – as drafted, under their BRI, in view of the specification, cover concepts performed in the human mind (evaluation, judgement, or opinion to map/correlate augmented graphs corresponding to shared/common individual graphs with shorter vectors, and to map/correlate augmented graphs corresponding to different/distinct individual graphs with longer vectors based on observed data in the augmented graphs).
The above limitations in the context of these claims encompass, inter alia, mapping/correlating augmented graphs corresponding to shared/common and different/distinct individual graphs with relatively short and long vectors based on observed data in the augmented graphs (evaluation/judgement/opinion to map/correlate augmented graphs corresponding to individual graphs with vectors based on observed data in the augmented graphs) (corresponding to mental processes which can be done mentally or by pen and paper).
Regarding the “individual graphs” and “augmented graphs”, aside from including nodes and data, no details of the graphs are recited and the graphs are recited at a high level of generality and can be constructed by hand with pen and paper. Thus, the claimed “individual graphs” and “augmented graphs”, under the BRI, in light of the specification, could be constructed by hand with pen and paper based on a reasonable amount of observed data.
Step 2A Prong Two Analysis: The judicial exceptions are not integrated into a practical application.
In particular, these claims recite, using respective similar language, the additional element of train the encoder to map augmented graphs – This additional element in the claims amounts to recitation of the words "apply it" (or an equivalent) or are mere instructions to implement an abstract idea or other exception on a computer, which does not integrate a judicial exception into a practical application. See MPEP 2106.05(f). In particular, “train the encoder to map augmented graphs” is simply generic training to perform the abstract idea of processing data to map/correlate data in augmented graphs and amounts to mere instructions to apply the exception (MPEP 2106.05(f)).
Merely asserting that a judicial exception is to be carried out on a generic computer (i.e., with the generically-recited “encoder”, modules and the “device comprising: a memory … and a processor” of base claim 1) cannot meaningfully integrate the judicial exception into a practical application. See MPEP § 2106.05(f).
Step 2B Analysis: The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above with respect to the integration of the abstract idea into a practical application, the additional elements of:
“train the encoder to map augmented graphs” is generic training to perform the abstract idea and amounts to no more than mere instructions to apply the exception (MPEP 2106.05(f)).
Mere instructions to apply the mental process electronically (i.e., with the generically-recited “encoder” and the “device comprising: a memory … and a processor” of base claim 1) do not amount to significantly more than the judicial exception. As noted above, merely asserting that a judicial exception is to be carried out on a generic computer cannot provide significantly more than the judicial exception. See MPEP § 2106.05(f).
Accordingly, at Step 2B, the additional element does not amount to significantly more than the judicial exception.
Regarding claims 7 and 15, these claims are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 7 is directed to a device as depending from claim 5, and claim 15 is directed to a method as depending from claim 13, thus the analysis for patent eligibilities of claims 5 and 13, of intervening claims 4 and 12, and of base claims 1 and 9 are incorporated herein.
Step 2A Prong 1: The claims both recite wherein a vector space for defining vectors corresponding to the plurality of augmented graphs is different from a vector space for defining the contrastive loss for the plurality of augmented graphs.
The additional limitation added by these claims merely limits the invention to a narrower abstract idea by merely reciting a type of data included in “a vector space for defining vectors”. This limitation does nothing to alter the fundamental nature of the claims as a mental process. This is because the additional limitation merely limits the invention to a narrower abstract idea by further narrowing what the “vector space for defining vectors” includes data that “is different from a vector space for defining the contrastive loss for the plurality of augmented graphs.”
Dependent claims 7 and 15, when analyzed as a whole, are not patent eligible under 35 U.S.C. 101 because the additional recited limitation fails to establish that the claims are not directed to an abstract idea.
Regarding the “augmented graphs”, aside from including nodes and data, no details of the graphs are recited and the graphs are recited at a high level of generality and can be constructed by hand with pen and paper. Thus, the claimed “augmented graphs”, under the BRI, in light of the specification, could be constructed by hand with pen and paper based on a reasonable amount of observed data.
Step 2A Prong Two Analysis: The judicial exceptions are not integrated into a practical application.
The claims do not recite any additional elements that integrate the abstract idea into a practical application or provide significantly more than the abstract idea, and thus the claims are subject-matter ineligible.
Step 2B Analysis: The claims do not recite additional elements that are sufficient to amount to significantly more than the judicial exception.
These claims are not patent eligible.
Regarding claim 8, this claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 8 is directed to a device as depending from claim 5, thus the analysis for patent eligibilities of claim 5, of intervening claim 4, and of base claim 1 is incorporated herein.
Step 2A Prong 1: The claim recites wherein the contrastive loss comprises two or more positive pairs defined based on vectors corresponding to the three or more augmented graphs - This limitation, as drafted, under its BRI, in view of the specification, covers concepts performed in the human mind (evaluation, judgement, or opinion to identify/determine a contrastive loss/difference by comparing/contrasting positive pairs of data items/values based on vectors corresponding to observed data in the augmented graphs) (corresponding to mental processes which can be done mentally or by pen and paper).
Regarding the “augmented graphs”, aside from including nodes and data, no details of the graphs are recited and the graphs are recited at a high level of generality and can be constructed by hand with pen and paper. Thus, the claimed “augmented graphs”, under the BRI, in light of the specification, could be constructed by hand with pen and paper based on a reasonable amount of observed data.
Step 2A Prong Two Analysis: The judicial exceptions are not integrated into a practical application.
In particular, this claim recites the additional element of wherein the processor is further configured to control the augmentation module to obtain three or more augmented graphs - This additional element in the claim amounts to recitation of the words "apply it" (or an equivalent) or are mere instructions to implement an abstract idea or other exception on a computer, which does not integrate a judicial exception into a practical application. See MPEP 2106.05(f).
Merely asserting that a judicial exception is to be carried out on a generic computer (i.e., with the generically-recited “module” and the “processor”) cannot meaningfully integrate the judicial exception into a practical application. See MPEP § 2106.05(f).
This claim also recites the additional element of obtain three or more augmented graphs for each of the plurality of individual graphs by augmenting each of the plurality of individual graphs by three or more different methods – This is an insignificant extra-solution activity that is not integrated into the claim as a whole and does not add a meaningful limitation to the above-noted mental processes specified in this claim. That is, “obtain three or more augmented graphs for each of the plurality of individual graphs by augmenting each of the plurality of individual graphs by three or more different methods” is mere data gathering (See MPEP § 2106.05(g)).
The claim does not recite any additional elements that integrate the abstract idea into a practical application or provide significantly more than the abstract idea, and thus the claim is subject-matter ineligible.
Step 2B Analysis: The claim does not recite additional elements that are sufficient to amount to significantly more than the judicial exception.
Mere instructions to apply the mental process electronically (i.e., with the generically-recited “module” and the “processor”) do not amount to significantly more than the judicial exception. As noted above, merely asserting that a judicial exception is to be carried out on a generic computer cannot provide significantly more than the judicial exception. See MPEP § 2106.05(f).
Moreover, receiving, communicating, forwarding and storing data are insignificant extra-solution activities that are well-understood, routine, and conventional. See MPEP2106.05(d)(II) (“The courts have recognized the following computer functions as well-understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity. i. Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); … iv. Storing and retrieving information in memory”) (citing OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015)) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network)". Therefore, the recitation of “obtain three or more augmented graphs for each of the plurality of individual graphs by augmenting each of the plurality of individual graphs by three or more different methods” are the well-understood, routine, conventional activities of receiving and transmitting data over a network, as discussed in MPEP § 2106.05(d).
Accordingly, at Step 2B, the additional element does not amount to significantly more than the judicial exception.
This claim is not patent eligible.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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 1-3, 9-11, 16-17 and 20 are rejected under 103 over non-patent literature Dong et al. ("Improving Sequential Recommendation with Attribute-augmented Graph Neural Networks." arXiv preprint arXiv:2103.05923 (March 2021), hereinafter “Dong”) in view of Brewer et al. (U.S. Patent Publication No. 2018/0060736 A1, hereinafter “Brewer”).
With respect to claim 1, Dong discloses the invention as claimed including obtain a plurality of individual graphs representing an access history of a user for a plurality of contents by a plurality of sessions (see, e.g., Abstract, “we propose an attribute-augmented graph neural network model named Murzim. Murzim takes as input the graphs constructed from the user-item interaction sequences and corresponding item attribute sequences.” and page 1, Sect. 1, “recommendation is to predict the next item that a user is most likely to interact with according to the user-item interaction sequence over a period of time in the past and then recommend the predicted item to the user. … scenarios include but are not limited to e-commerce platforms where products are recommended based on the user click records in the recent period, and video streaming platforms where videos are recommended to users based on their historical watching records. … records in a user-item interaction sequence are sorted” [i.e., obtain graphs representing a user’s content item access history/interactions for sessions/sequences]);
generate an integrated graph, in which the plurality of individual graphs are integrated, based on a connection relationship between nodes included in the plurality of individual graphs and a number of times each connection between the nodes is repeated (see, e.g., pages 2, Sect. 1, “sequential recommendation model Murzim. Based on gated GNNs [graph neural networks], Murzim adopts attention mechanisms to integrate information from the node level and the sequence level, and fuses the influence of item attributes on the semantics implied in user-item interaction sequences into the recommendation results.” and 4, Sect. 3, “given the user-item interaction sequence set S, we use V = {v1; v2;…; v|V|} to denote the set consisting of all unique items involved in all the sequences, P” [i.e., generate an integrated graph integrating individual graphs based on node level connections and sequences/number of times connections are repeated]);
obtain a plurality of augmented graphs by augmenting each of the plurality of individual graphs based on the integrated graph (see, e.g., FIG. 1 – reproduced below, depicting “Update” and “Attentive Node Aggregation” of plurality of individual “Sequence Graphs” to obtain augmented graphs:
PNG
media_image1.png
200
400
media_image1.png
Greyscale
and pages 2, Sect. 1, “attention mechanisms to integrate information from the node level and the sequence level, and fuses the influence of item attributes on the semantics implied in user-item interaction sequences” and 4, Sect. 4, “Then, we use gated GNN based on GRU to update the embeddings of nodes in the graphs. After obtaining the embeddings of nodes, we aggregate them to get the embeddings of the sequences.” [i.e., obtain augmented/updated graphs by augmenting each of the plurality of individual sequence graphs based on the integrated graph from an attention mechanism]); and
train the neural network model to provide recommended content based on the plurality of augmented graphs (see, e.g., Abstract, “we propose an attribute-augmented graph neural network model named Murzim. … Murzim can capture user preference patterns, generate embeddings for user-item interaction sequences, and then generate recommendations through next-item prediction.” and pages 7, Sect. 4.4, “ϴ is the set of all trainable parameters of the model. We train the model by optimizing L through the gradient descent method.” and 10, Sect. 5.3, “We have deployed Murzim to … generate recommendation for users based on their viewing sequences.” [i.e., train the graph neural network model to provide recommendations/recommended content based on the augmented graphs]).
Although Dong substantially discloses the claimed invention, Dong is not relied on for explicitly disclosing an electronic device comprising: a memory storing a neural network model; and a processor configured to: <perform operations>.
However, in the same field, analogous art Brewer teaches an electronic device comprising: a memory storing a neural network model; and a processor configured to: <perform operations> (see, e.g., paragraphs 33, “model generating module 206 is configured to train a model for generating latent representations. In some embodiments, the model is a single layer recurrent neural network” [i.e., a neural network model] and 81, “computer system 700 includes a processor 702, a cache 704, and one or more executable modules and drivers, stored on a computer-readable medium, directed to the processes and features described herein. … Additionally, the computer system 700 includes … A system memory 714” and 83-84, “mass storage 718 provides permanent storage for the data and programming instructions to perform the above-described processes and features implemented by the respective computing systems identified above, whereas the system memory 714 (e.g., DRAM) provides temporary storage for the data and programming instructions when executed by the processor 702.”, “computer system 700 may include additional components, such as additional processors, storage devices, or memories.” [i.e., a system and device including a memory/storage storing a neural network model and a processor configured to perform operations]).
Dong and Brewer are analogous art because they are both directed to implementing and training neural network models and machine learning models for recommending content items (see, e.g., Dong, Abstract and page 8, and Brewer, Abstract, and paragraphs 30, 33 and 47).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Dong to incorporate the teachings of Brewer to provide a “networking system 630 [that] generates and maintains the "social graph" comprising a plurality of nodes interconnected by a plurality of edges. … types of nodes include users, non-person entities, content items, web pages, groups, activities, messages”, “when a first user identifies a second user as a friend, an edge in the social graph is generated connecting a node representing the first user and a second node representing the second user. As various nodes relate or interact with each other, the social networking system 630 modifies edges connecting the various nodes to reflect the relationships and interactions.” and “networking system 630 maintains data about objects with which a user may interact (see, e.g., Brewer, paragraphs 62-63 and 67). Doing so would have allowed Dong to use Brewer’s “networking system 630 [that] is also capable of linking a variety of entities.” Where the “networking system 630 enables users to interact with each other as well as external systems 620 or other entities through an API, a web service, or other communication channels”, as suggested by Brewer (See, e.g., Brewer, paragraph 62).
With respect to independent claim 9, claim 9 is substantially similar to claim 1 and therefore is rejected on the same grounds as claim 1, discussed above. In particular, claim 9 is a method claim with steps that correspond to the device operations of claim 1.
Dong further discloses a method for controlling an electronic device using a neural network model (see, e.g., Abstract, “we propose an attribute-augmented graph neural network model named Murzim. Murzim takes as input the graphs constructed from the user-item interaction sequences and corresponding item attribute sequences. … Murzim has been deployed in MX Player, one of India's largest streaming platforms” and pages 1-2, Sect. 1, “The target scenarios include but are not limited to e-commerce platforms where products are recommended based on the user click records in the recent period, and video streaming platforms where videos are recommended to users based on their historical watching records.”, “We present a reasonable method to construct attribute sequences from user-item interaction sequences and attribute graphs from attribute sequences.” [i.e., a technique/method for controlling an electronic device in a streaming platform using a neural network model]).
With respect to independent claim 16, claim 16 is substantially similar to claim 1 and therefore is rejected on the same grounds as claim 1, discussed above. In particular, claim 16 is a computer readable storage medium with operations that correspond to the device operations of claim 1.
Although Dong substantially discloses the claimed invention, Dong is not relied on for explicitly disclosing a non-transitory computer readable storage medium storing program instructions that are executable by a processor, wherein the program instructions, when executed by the processor, are configured to control a device to: <perform operations>.
However, in the same field, analogous art Brewer teaches a non-transitory computer readable storage medium storing program instructions that are executable by a processor, wherein the program instructions, when executed by the processor, are configured to control a device to: <perform operations> (see, e.g., paragraphs 81, “computer system 700 includes a processor 702, a cache 704, and one or more executable modules and drivers, stored on a computer-readable medium, directed to the processes and features described herein.” and 86, “The foregoing modules may be realized by hardware, executable modules stored on a computer-readable medium (or machine-readable medium), or a combination of both. For example, the modules may comprise a plurality or series of instructions to be executed by a processor in a hardware system, such as the processor 702.” [i.e., a computer-readable storage medium storing executable instructions when executed by a processor, are configured to perform operations]).
Dong and Brewer are analogous art because they are both directed to implementing and training neural network models and machine learning models for recommending content items (see, e.g., Dong, Abstract and page 8, and Brewer, Abstract and paragraphs 30, 33 and 47).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Dong to incorporate the teachings of Brewer to provide a “networking system 630 [that] generates and maintains the "social graph" comprising a plurality of nodes interconnected by a plurality of edges. … types of nodes include users, non-person entities, content items, web pages, groups, activities, messages”, “when a first user identifies a second user as a friend, an edge in the social graph is generated connecting a node representing the first user and a second node representing the second user. As various nodes relate or interact with each other, the social networking system 630 modifies edges connecting the various nodes to reflect the relationships and interactions.” and “networking system 630 maintains data about objects with which a user may interact (see, e.g., Brewer, paragraphs 62-63 and 67). Doing so would have allowed Dong to use Brewer’s “networking system 630 [that] is also capable of linking a variety of entities.” where the “networking system 630 enables users to interact with each other as well as external systems 620 or other entities through an API, a web service, or other communication channels”, as suggested by Brewer (See, e.g., Brewer, paragraph 62).
With respect to independent claim 17, Dong discloses the invention as claimed including obtain a plurality of individual graphs based on access history of a user for a plurality of contents (see, e.g., Abstract, “we propose an attribute-augmented graph neural network model named Murzim. Murzim takes as input the graphs constructed from the user-item interaction sequences and corresponding item attribute sequences.” and page 1, Sect. 1, “recommendation is to predict the next item that a user is most likely to interact with according to the user-item interaction sequence over a period of time in the past and then recommend the predicted item to the user. … scenarios include but are not limited to e-commerce platforms where products are recommended based on the user click records in the recent period, and video streaming platforms where videos are recommended to users based on their historical watching records. … records in a user-item interaction sequence are sorted” [i.e., obtain graphs representing a user’s content item access history/interactions in sequences]);
obtain a plurality of vectors based on the plurality of individual graphs (see, e.g., FIG. 1 – reproduced below, depicting “Attentive Node Aggregation” of plurality of individual “Sequence Graphs” to obtain vectors S0 S1 of “Sequence Embeddings”:
PNG
media_image1.png
200
400
media_image1.png
Greyscale
and pages 5, Sect. 4.2, “We get the initial item embeddings through the embedding look-up operations from a trainable matrix with dimension d x |V|. The d-dimensional vector vi is used to represent the embedding of the i-th item.” and 7, Sect. 4.4. “In formula (11), y
∈
R|V| is the one-hot encoding vector corresponding to the ground truth item.” [i.e., obtain vectors v based on the individual sequence graphs]);
identify recommended content by controlling a neural network model based on the plurality of vectors (see, e.g., FIG. 1 – reproduced above – depicting “Ranking Scores” for identifying recommended content by controlling the Murzim graph neural network model based on vectors S0 S1 in “Sequence Embeddings” and Abstract, “we propose an attribute-augmented graph neural network model named Murzim. … Murzim can capture user preference patterns, generate embeddings for user-item interaction sequences, and then generate recommendations through next-item prediction.” and pages 5, Sect. 4.2, “We get the initial item embeddings through the embedding look-up operations from a trainable matrix with dimension d x |V|. The d-dimensional vector vi is used to represent the embedding of the i-th item.” and 10, Sect. 5.3, “We have deployed Murzim to … generate recommendation for users based on their viewing sequences.” [i.e., based on the vectors, identify recommended content by controlling a graph neural network model]);
obtain an integrated graph based on each of the plurality of individual graphs (see, e.g., pages 2, Sect. 1, “sequential recommendation model Murzim. Based on gated GNNs [graph neural networks], Murzim adopts attention mechanisms to integrate information from the node level and the sequence level, and fuses the influence of item attributes on the semantics implied in user-item interaction sequences into the recommendation results.” and 4, Sect. 3, “given the user-item interaction sequence set S, we use V = {v1; v2;…; v|V|} to denote the set consisting of all unique items involved in all the sequences, P” [i.e., obtain an integrated graph based on individual sequence graphs - based on node level connections and sequences in the individual sequence graphs]);
obtain a plurality of augmented graphs based on each of the integrated graph[s] (see, e.g., FIG. 1 – reproduced above, depicting “Update” and “Attentive Node Aggregation” of plurality of individual “Sequence Graphs” to obtain augmented graphs, and pages 2, Sect. 1, “attention mechanisms to integrate information from the node level and the sequence level, and fuses the influence of item attributes on the semantics implied in user-item interaction sequences” and 4, Sect. 4, “Then, we use gated GNN based on GRU to update the embeddings of nodes in the graphs. After obtaining the embeddings of nodes, we aggregate them to get the embeddings of the sequences.” [i.e., obtain augmented/updated graphs by augmenting each of the plurality of individual sequence graphs based on the integrated graph from the attention mechanism]);
obtaining a plurality of augmented vectors based on the plurality of augmented graphs (see, e.g., e.g., FIG. 1 – reproduced above – depicting obtaining updated/augmented S0 S1 vectors in “Sequence Embeddings” based on updated/augmented “Sequence Graphs”, and page 5, Sect. 4.2, “We get the initial item embeddings through the embedding look-up operations from a trainable matrix with dimension d x |V|. The d-dimensional vector vi is used to represent the embedding of the i-th item.” [i.e., obtain updated/augmented vectors based on the augmented sequence graphs]); and
train the neural network model to provide recommended content based on the plurality of augmented graphs (see, e.g., pages 7, Sect. 4.4, “ϴ is the set of all trainable parameters of the model. We train the model by optimizing L through the gradient descent method.” [i.e., train the graph neural network model based on the augmented vectors]).
Although Dong substantially discloses the claimed invention, Dong is not relied on for explicitly disclosing a device comprising: a memory; and a processor configured to execute instructions stored in the memory to: <perform operations>.
However, in the same field, analogous art Brewer teaches a device comprising: a memory; and a processor configured to execute instructions stored in the memory to: <perform operations> (see, e.g., paragraphs 81, “computer system 700 includes a processor 702, a cache 704, and one or more executable modules and drivers, stored on a computer-readable medium, directed to the processes and features described herein. … Additionally, the computer system 700 includes … A system memory 714” and 83-84, “mass storage 718 provides permanent storage for the data and programming instructions to perform the above-described processes and features implemented by the respective computing systems identified above, whereas the system memory 714 (e.g., DRAM) provides temporary storage for the data and programming instructions when executed by the processor 702.”, “computer system 700 may include additional components, such as additional processors, storage devices, or memories.” [i.e., a system and device including a memory/storage storing a executable instructions and a processor configured to execute the instructions to perform operations]).
Dong and Brewer are analogous art because they are both directed to implementing and training neural network models and machine learning models for recommending content items (see, e.g., Dong, Abstract and page 8, and Brewer, Abstract and paragraphs 30, 33 and 47).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Dong to incorporate the teachings of Brewer to provide a “networking system 630 [that] generates and maintains the "social graph" comprising a plurality of nodes interconnected by a plurality of edges. … types of nodes include users, non-person entities, content items, web pages, groups, activities, messages”, “when a first user identifies a second user as a friend, an edge in the social graph is generated connecting a node representing the first user and a second node representing the second user. As various nodes relate or interact with each other, the social networking system 630 modifies edges connecting the various nodes to reflect the relationships and interactions.” and “networking system 630 maintains data about objects with which a user may interact (see, e.g., Brewer, paragraphs 62-63 and 67). Doing so would have allowed Dong to use Brewer’s “networking system 630 [that] is also capable of linking a variety of entities.” where the “networking system 630 enables users to interact with each other as well as external systems 620 or other entities through an API, a web service, or other communication channels”, as suggested by Brewer (See, e.g., Brewer, paragraph 62).
Regarding claims 2, 10 and 20, as discussed above, Dong in view of Brewer teaches the devices of claims 1 and 17, and the method of claim 9.
Dong further discloses obtain at least one augmented graph of the plurality of augmented graphs by adding at least one node included in the integrated graph to a plurality of nodes included in each of the plurality of individual graphs (see, e.g., FIG. 1 – reproduced above, depicting “Update” and “Attentive Node Aggregation” of plurality of individual “Sequence Graphs” to obtain at least one augmented graph by aggregating and adding at least one node to nodes of the individual sequence graphs and pages 2, Sect. 1, “attention mechanisms to integrate information from the node level and the sequence level, and fuses the influence of item attributes on the semantics implied in user-item interaction sequences”, 4, Sect. 4, “the embeddings of nodes in the graphs. After obtaining the embeddings of nodes, we aggregate them to get the embeddings of the sequences.” and 5-6, Sects. 4.1 and 4.3, “we add a directed edge between the corresponding nodes in the graph”, “We first aggregate the embeddings of the nodes in different graphs to get the embeddings” [i.e., obtain at least one augmented graph by aggregating and adding at least one node to nodes of the individual graphs]).
Regarding claims 3 and 11, as discussed above, Dong in view of Brewer teaches the device of claim 1 and the method of claim 9.
Dong further discloses obtain at least one augmented graph of the plurality of augmented graphs by changing at least one node of a plurality of nodes included in each of the plurality of individual graphs to at least one node included in the integrated graph (see, e.g., FIG. 1 – reproduced above, depicting “Update” of plurality of individual “Sequence Graphs” to obtain at least one augmented graph by changing/updating at least one node included in “Attentive Node Aggregation”/nodes the individual sequence graphs and pages 2, Sect. 1, “attention mechanisms to integrate information from the node level and the sequence level, and fuses the influence of item attributes on the semantics implied in user-item interaction sequences”, 4, Sect. 4, “Then, we use gated GNN based on GRU to update the embeddings of nodes in the graphs. After obtaining the embeddings of nodes, we aggregate them to get the embeddings of the sequences.” and 6, Sect. 4.2, “Let ei represent the embedding of node i on the graph (item graph or attribute graph), and then we propagate the information between nodes … In the subsequent step, mi
∈
R2dx1 is used as the input of the GRU to updated [sic – update] the embedding of node i” [i.e., obtain at least one augmented graph by changing/updating at least one node included in the individual sequence graphs]).
Claims 4-5, 7, 12-13, 15 and 18-19 are rejected under 103 over non-patent literature Dong et al. ("Improving Sequential Recommendation with Attribute-augmented Graph Neural Networks." arXiv preprint arXiv:2103.05923 (March 2021), hereinafter “Dong”) in view of Brewer et al. (U.S. Patent Publication No. 2018/0060736 A1, hereinafter “Brewer”) and further in view of non-patent literature XIE, XU et al. ("Contrastive Learning for Sequential Recommendation", SIGIR '21, July 11-15, 2021, Montreal, Canada, arXiv:2010.14395v2 [cs.IR], February 28, 2021, cited in Applicant’s IDS filed 5/24/2023, hereinafter “Xie”).
Regarding claims 4 and 12, as discussed above, Dong in view of Brewer teaches the device of claim 1 and the method of claim 9.
Although Dong in view of Brewer substantially teaches the claimed invention and Brewer discloses that “The model itself is trained using contrastive divergence.” (see, e.g., paragraph 34), Dong in view of Brewer is not relied on to teach train the neural network model based on a contrastive loss for the plurality of augmented graphs.
In the same field, analogous art Xie teaches train the neural network model based on a contrastive loss for the plurality of augmented graphs (see, e.g., FIG. 1 – depicting framework where “Two data augmentation methods, 𝑎𝑖 and 𝑎𝑗 , are sampled from the same augmentation set A. They are applied to each user’s sequence and then we can obtain two correlated views of each sequence. A shared embedding layer and the user representation model 𝑓(・) transform the original and augmented sequences to the latent space where the contrastive loss and recommendation loss are applied.” and pages 2, Sect. 2.1, “each user sequence can be constructed as a graph” and 3, Sects. 3, 3.2 and 3.2.2, “we propose how to train the user representation model via a multi-task learning framework”, “we seek to explore applying contrastive learning algorithms to sequential recommendation task to obtain a powerful user representation model. The framework comprises three major components, including a stochastic data augmentation module, a user representation encoder, and a contrastive loss function.” [i.e., contrastive loss for augmentation graphs], “We utilize a neural network as a user representation encoder to extract information from the augmented sequences.” [i.e., train the neural network based on contrastive loss for the augmented graphs]).
Dong, Xie and Brewer are analogous art because they are each directed to implementing and training neural network models and machine learning models for recommending content items (see, e.g., Dong, Abstract and page 8, Sect. 5, Xie, Abstract and pages 1-2, Sect. 1, and Brewer, Abstract and paragraphs 30, 33 and 47).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Dong in view of Brewer to incorporate the teachings of Xie to provide a “multi-task model called Contrastive Learning for Sequential Recommendation (CL4SRec)” that “takes advantage of the traditional next item prediction task but also utilizes the contrastive learning framework to derive self-supervision signals from the original user behavior sequences.” (See, e.g., Xie, Abstract). Doing so would have allowed Dong in view of Brewer to employ Xie’s Contrastive Learning for Sequential Recommendation (CL4SRec) framework to “extract more meaningful user patterns and further encode the user representation effectively.” where CL4SRec “achieves state-of-the-art performance over existing baselines by inferring better user representations”, as suggested by Xie (See, e.g., Xie, Abstract).
Regarding claims 5 and 13, as discussed above, Dong in view of Brewer and further in view of Xie teaches the device of claim 4 and the method of claim 12.
Although Dong in view of Brewer substantially teaches the claimed invention and Dong discloses “a recommendation model based on an encoder-decoder structure, which designs an RNN with an attention mechanism in the encoder to capture the user's sequential behaviors and main purpose, and predicts the next item in the decoder.” (see, e.g., Dong, page 3, Sect. 2) and Brewer discloses that “the model can operate as an auto-encoder that produces a latent representation for each entity.” (see, e.g., paragraph 33), Dong in view of Brewer is not relied on to teach an encoder configured to obtain vectors corresponding to the plurality of individual graphs and vectors corresponding to the plurality of augmented graphs,
a prediction module configured to obtain recommended content information corresponding to each of the plurality of individual graphs,
an augmentation module configured to obtain the plurality of augmented graphs by augmenting each of the plurality of individual graphs, and
a contrastive loss obtaining module configured to obtain the contrastive loss for the plurality of augmented graphs.
In the same field, analogous art Xie teaches an encoder configured to obtain vectors corresponding to the plurality of individual graphs and vectors corresponding to the plurality of augmented graphs (see, e.g., FIG. 1 – depicting framework including “encoder” and pages 3, Sects. 3.2 and 3.2.2, “The framework comprises three major components, including … a user representation encoder”, “User Representation Encoder. We utilize a neural network as a user representation encoder to extract information from the augmented sequences. With this encoder, we can obtain meaningful user representations from their augmented sequences” and 5, Sects. 3.4, 3.4.1, “model in this paper, which applies a unidirectional Transformer encoder and has achieved promising results in the sequential recommendation task”, “The Transformer encoder utilizes an item embedding matrix 𝑬 ∈ R|V|×𝑑 to project high dimensional one-hot item representations to low dimensional dense vectors”),
a prediction module configured to obtain recommended content information corresponding to each of the plurality of individual graphs (see, e.g., pages 5-6, Sects. 3.4.2 and 3.4.5, “In sequential recommendation, we can only utilize the information before the time step 𝑡 when we predict the next item”, “our task is to predict the item at the time step |𝑠𝑢 | + 1 for each user 𝑢, we set the final representation for user 𝑢 as her preference vector at time |𝑠𝑢 | + 1” [i.e., prediction module obtains recommended content item corresponding to graphs]),
an augmentation module configured to obtain the plurality of augmented graphs by augmenting each of the plurality of individual graphs (see, e.g., page 3, Sects. 3.2-3.2.1, “The framework comprises three major components, including a stochastic data augmentation module”, “Data Augmentation Module. A stochastic data augmentation module is employed to transform each data sample randomly into two correlated instances. … we apply two randomly sampled augmentation methods (𝑎𝑖 ∈ A and 𝑎𝑗 ∈ A) to each user’s historical behaviors sequence 𝑠𝑢, and obtain two views of the sequence” [augmentation module augments individual graphs/sequences]), and
a contrastive loss obtaining module configured to obtain the contrastive loss for the plurality of augmented graphs (see, e.g., FIG. 1 – depicting framework including “contrastive loss” where “the user representation model 𝑓 (・) transform the original and augmented sequences to the latent space where the contrastive loss and recommendation loss are applied” and pages 3, Sect. 3.2 , “we seek to explore applying contrastive learning algorithms to sequential recommendation task to obtain a powerful user representation model. The framework comprises three major components, including … a contrastive loss function.” and 4, Sect. 3.2.3, “Contrastive Loss Function. Finally, a contrastive loss function is applied to distinguish whether the two representations are derived from the same user historical sequence. To achieve this target, the contrastive loss learns to minimize the difference between differently augmented views of the same user historical sequence and maximize the difference between the augmented sequences derived from different users.” [i.e., framework includes a contrastive loss function/module to obtain contrastive loss for the augmented sequences/graphs]).
The motivation to combine Dong, Brewer and Xie is the same as discussed above with respect to claims 4 and 12.
Regarding claims 7 and 15, as discussed above, Dong in view of Brewer and further in view of Xie teaches the device of claim 5 and the method of claim 13.
Although Dong in view of Brewer substantially teaches the claimed invention, Dong in view of Brewer is not relied on to teach wherein a vector space for defining vectors corresponding to the plurality of augmented graphs is different from a vector space for defining the contrastive loss for the plurality of augmented graphs.
In the same field, analogous art Xie teaches wherein a vector space for defining vectors corresponding to the plurality of augmented graphs is different from a vector space for defining the contrastive loss for the plurality of augmented graphs (see, e.g., pages 4, Sect. 3.2.3, “a contrastive loss function is applied to distinguish whether the two representations are derived from the same user historical sequence. To achieve this target, the contrastive loss learns to minimize the difference between differently augmented views of the same user historical sequence and maximize the difference between the augmented sequences derived from different users.” and 5, Sects. 3.4 and 3.4.2, “encoder utilizes an item embedding matrix 𝑬 ∈ R|V|×𝑑 to project high dimensional one-hot item representations to low dimensional dense vectors. In addition, to represent the position information of sequence”, “extract the information from different subspaces at each position, here we adopt the multi-head self-attention instead of a single attention function. It first utilizes different linear projections to project the input representations into ℎ subspaces.” [i.e., vector subspace for vectors corresponding to augmented sequences/graphs is different from vector subspace defining the contrastive loss]).
The motivation to combine Dong, Brewer and Xie is the same as discussed above with respect to claims 4 and 12.
Regarding claim 18, as discussed above, Dong in view of Brewer teaches the device of claim 17.
Although Dong in view of Brewer substantially teaches the claimed invention, Dong in view of Brewer is not relied on to teach wherein the processor is further configured to execute the instructions stored in the memory to obtain a main loss by comparing the recommended content identified based on the plurality of vectors with label information stored in the memory.
In the same field, analogous art Xie teaches wherein the processor is further configured to execute the instructions stored in the memory to obtain a main loss by comparing the recommended content identified based on the plurality of vectors with label information stored in the memory (see, e.g., page 6, Sects. 3.5 and 4.1.2, “unlabeled raw data to enhance the performance of sequential recommendation, we adopt a multi-task strategy where the main sequence prediction task and the additional contrastive learning task are jointly optimized. The total loss is a linear weighted sum as follows: Ltotal = Lmain + 𝜆Lcl. (14) We adopt the negative log likelihood with sampled softmax as the main loss for each user 𝑢 at each time step 𝑡 + 1”, “For each user, we hold out the last interacted item as the test data and utilize the item just before the last as the validation data.” [i.e., obtain a main loss/Lmain by comparing recommended content items with stored label information/validation data]).
Dong, Xie and Brewer are analogous art because they are each directed to implementing and training neural network models and machine learning models for recommending content items (see, e.g., Dong, Abstract and page 8, Sect. 5, Xie, Abstract and pages 1-2, Sect. 1, and Brewer, Abstract and paragraphs 30, 33 and 47).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Dong in view of Brewer to incorporate the teachings of Xie to provide a “multi-task model called Contrastive Learning for Sequential Recommendation (CL4SRec)” that “takes advantage of the traditional next item prediction task but also utilizes the contrastive learning framework to derive self-supervision signals from the original user behavior sequences.” (See, e.g., Xie, Abstract). Doing so would have allowed Dong in view of Brewer to employ Xie’s Contrastive Learning for Sequential Recommendation (CL4SRec) framework to “extract more meaningful user patterns and further encode the user representation effectively.” where CL4SRec “achieves state-of-the-art performance over existing baselines by inferring better user representations”, as suggested by Xie (See, e.g., Xie, Abstract).
Regarding claim 19, as discussed above, Dong in view of Brewer and further in view of Xie teaches the device of claim 18.
Although Dong in view of Brewer substantially teaches the claimed invention, Dong in view of Brewer is not relied on to teach wherein the processor is further configured to execute the instructions stored in the memory to train the neural network model based on the main loss.
In the same field, analogous art Xie teaches wherein the processor is further configured to execute the instructions stored in the memory to train the neural network model based on the main loss (see, e.g., pages 3, Sect. 3.2.2, “We utilize a neural network as a user representation encoder to extract information from the augmented sequences” and 6, Sects. 3.5 and 4.1.2, “Multi-task Training To leverage the self-supervised signals derived from the unlabeled raw data to enhance the performance of sequential recommendation, we adopt a multi-task strategy where the main sequence prediction task and the additional contrastive learning task are jointly optimized. The total loss is a linear weighted sum as follows: Ltotal = Lmain + 𝜆Lcl. (14) We adopt the negative log likelihood with sampled softmax as the main loss for each user 𝑢 at each time step 𝑡 + 1”, “For each user, we hold out the last interacted item as the test data and utilize the item just before the last as the validation data. The remaining items are used for training.” [i.e., train the neural network model based on the main loss/Lmain])
The motivation to combine Dong, Brewer and Xie is the same as discussed above with respect to claim 18.
Conclusion
The prior art made of record, listed on form PTO-892, and not relied upon, is considered pertinent to applicant's disclosure; and all references generally relate to techniques, methods and systems for neural network models and knowledge graph exploration.
For example, non-patent literature Zhang, et al. ("Personalized Graph Neural Networks with Attention Mechanism for Session-Aware Recommendation." arXiv preprint arXiv:1910.08887 v4 (Jan 2021), hereinafter “Zhang”) discloses “a novel method, named Personalized Graph Neural Networks with Attention Mechanism (A-PGNN) for brevity. A-PGNN mainly consists of two components: one is Personalized Graph Neural Network (PGNN), which is used to extract the personalized structural information in each user behavior graph, compared with the traditional Graph Neural Network (GNN) model, which considers the role of the user when the node embeddding [sic - embedding] is updated. The other is Dot-Product Attention mechanism, which draws on the Transformer net to explicitly model the effect of historical sessions on the current session. Extensive experiments conducted on two real-world data sets show that A-PGNN evidently outperforms the state-of-the-art personalized session-aware recommendation methods.” (see, Abstract).
The examiner requests, in response to this office action, support be shown for language added to any original claims on amendment and any new claims. That is, indicate support for newly added claim language by specifically pointing to page(s) and line no(s) in the specification and/or drawing figure(s). This will assist the examiner in prosecuting the application.
When responding to this office action, Applicant is advised to clearly point out the patentable novelty which he or she thinks the claims present, in view of the state of the art disclosed by the reference cited or the objections made. He or she must also show how the amendments avoid such references or objections See 37 CFR 1.111 (c).
Any inquiry concerning this communication or earlier communications from the examiner should be directed to RANDY K BALDWIN whose telephone number is (571)270-5222. The examiner can normally be reached on Mon - Fri 9:00-6:00.
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 on 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.
/RANDALL K. BALDWIN/Primary Examiner, Art Unit 2125
1 As indicated in the section 112(b) rejections of these claims above, “a short distance on a vector space" is being interpreted as any distance measurement or amount that is less than the recited “long distance on the vector space”.
Prosecution Insights