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 .
Response to Amendment
The office action is responsive to the amendment filed on 09/17/2025. As directed by the amendments claims 1, 4, 6, 7, 9 and 12 have been amended. Claims 1-20 are pending for examination.
Response to Arguments
Regarding the 35 U.S.C § 112 Rejection:
Applicant’s arguments, see pg. 10, filed on 09/17/2025, with respect to claims 1-20 being rejected under 35 U.S.C § 112 have been fully considered and are persuasive. The rejection of claims 1-20 under 35 U.S.C § 112 has been withdrawn.
Regarding the 35 U.S.C § 101 Rejection:
Applicant's further arguments see pg. 10-11 filed 09/17/2025, have been fully considered but they are not persuasive.
APPLICANT ARGUMENT:
Applicant argues, “The present application relates to the joint training of a plurality of models to separated features (representations) from training data. The training process of the model cannot be accomplished by a human thought process, but must rely on the functionality of a computer. A human being is also unable to separate the category-dependent representation and the category- independent representation by manual means, as these two features are not simply segmented from the data, but are extracted from the data by the model after learning. Applicant therefore asserts claim 1 of this application is eligible because it as a whole integrates alleged abstract idea into a practical application, i.e., automatic recommendation. Furthermore, the claim 1 of this application, for example, amounts to significantly more than any alleged abstract idea by improving efficiency and diversity of recommendations. Other claims are also patent-eligible for similar reasons. Reconsideration and withdrawal of the rejections is respectfully requested”.
EXAMINER RESPONSE: Examiner respectfully disagree, applicant argument is not persuasive. First to clarify, as presented in the Non-Final Office Action dated 06/17/2025 the training process of claim 1 was not view as an abstract idea (e.g., mental process) under STEP 2A Prong 1, rather examiner clearly stated the training process of the model was directed to using computers or other machinery merely as a tool to perform an existing process under STEP 2A Prong 2 and STEP 2B (see pg. 12-16 Non-Final Office Action dated 06/17/2025). Second, the claim language as presented differ from applicant argument. Amended claim 1 does not teach the category-dependent representation and the category- independent representation been separated. Rather, amended claim 1 teaches “processing data for training by using a recommendation model to obtain a category- independent representation and a category-dependent representation...” which is an abstract idea being implemented on a generic computer (i.e., using a recommendation model) (see MPEP 2106.05(f)).
Furthermore, amended claim 1 as presented does not integrate into a practical application under the second prong of the two-prong analysis since the claimed invention do not improves the functioning of a computer or improves another technology or technical field. Rather the claim recites additional element of:
A training method of a recommendation model, comprising:
...by using a recommendation model...
inputting the data for training into the recommendation model to obtain an output representation, wherein each piece of the output representation comprises the category-independent representation and the category-dependent representation,...
...the category-independent representation is for extracting a representation that is common to all categories from a user feature and an article feature, the category-dependent representation is for determining the categories that the user is interested in from the user feature and the article feature, each piece of the data for training comprises a feature of the user and a feature of an article, the feature of the user and the feature of the article of the piece of the data for training are input into the recommendation model simultaneously, and each piece of the data for training is pre-marked with recommendation information and a category of the article, and the recommendation information indicates whether a feedback on the article is provided by the user;
...by using a discriminator ...processed by the discriminator...
...by using a first mapping model ....by using a second mapping model...
training the recommendation model and the discriminator according to training targets comprising the category-independent representation not corresponding to any one of the plurality of categories, the category-dependent representation corresponding to the pre-marked category, and the prediction result matching with the pre-marked recommendation information, wherein the recommendation model is configured to determine whether to recommend an alternative article to a target user,...
...and data corresponding to determined articles recommended for the target user are sent to a terminal device of the target user for display
That merely recites the words "apply it" (or an equivalent) with the judicial exception, as discussed in MPEP § 2106.05(f), adds insignificant extra-solution activity to the judicial exception, as discussed in MPEP § 2106.05(g) and generally links the use of a judicial exception to a particular technological environment or field of use, as discussed in MPEP § 2106.05(h) which the courts have identified such limitations do not integrate a judicial exception into a practical application (see MPEP 2106.04(d)(I)).
Lastly, regarding the arguments “claim 1 of this application, for example, amounts to significantly more than any alleged abstract idea by improving efficiency and diversity of recommendation” examiner respectfully disagree. The amended claims as presented include limitations that the courts have identified not to be enough to qualify as “significantly more” when recited in a claim with a judicial exceptions this includes: adding the words “apply it” (or equivalent) with the judicial exception; simply appending well-understood, routine, conventional activities previously known to the industry, specified at high level of generality, to the judicial exception; adding insignificant extra solution activity to the judicial exception and generally linking the use of the judicial exception to a particular environment or field of use (see MPEP 2106.05 (I)(A)).
Therefore for the above reason, claims 1-20 are not directed to patent-eligible subject matter under 35 U.S.C § 101.
Regarding the 35 U.S.C § 103:
Applicant’s arguments with respect to claims 1-20 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
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.
Step 1:
Claims 1-8 are method type claims for recommendation model training. Claims 9-11 are method claims for article recommendation. Claims 12-16 are a training apparatus for a recommendation model type claims. Claims 17-18 are an article recommending apparatus type claims. Claim 19 is a computer program product claim for recommendation model training. Claim 20 is a computer program product claim for article recommendation. Therefore, claims 1-20 are directed to either a process, machine, manufacture, or composition of matter.
Regarding claim 1: 2A Prong 1:
...processing data for training ...to obtain a category- independent representation and a category-dependent representation, comprising:... (mental process – of processing data for training to obtain category- independent representation and a category-dependent representation can be performed by the human mind with the help of pen and paper (e.g., evaluation and judgment)).
processing the category-independent representation ...to obtain a first discrimination result corresponding to the category-independent representation, the first discrimination result indicating a correlation between the category-independent representation processed by the discriminator and a plurality of categories, and processing the category-dependent representation ...to obtain a second discrimination result corresponding to the category-dependent representation, the second discrimination result indicating a correlation between the category-dependent representation ... and the plurality of categories, wherein the processing the category-independent representation and the processing the category-dependent representation are performed separately; (mental process – of processing the category-independent representation and the processing the category-dependent representation separately to obtain discriminations results can be performed by the human mind with the help of pen and paper (e.g., evaluation and judgment)).
determining a prediction result according to at least one of the category-independent representation or the category-dependent representation, comprising: processing the category- independent representation ...to obtain a first prediction result, or processing the category-independent representation and the category-dependent representation ...to obtain a second prediction result; and (mental process – of determining a prediction result according to at least one of the category-independent representation or the category-dependent representation can be performed by the human mind with the help of pen and paper. For example, a person could manually evaluate a category-independent representation and then process the category-independent representation to obtain a first prediction result or manually evaluate category-independent representation and a category-dependent representation, and then process the representations to obtain a second prediction result (e.g., evaluation and judgment)).
2A Prong 2: This judicial exception is not integrated into a practical application.
Additional elements:
A training method of a recommendation model, comprising: (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
...by using a recommendation model... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
inputting the data for training into the recommendation model to obtain an output representation, wherein each piece of the output representation comprises the category-independent representation and the category-dependent representation,... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
...the category-independent representation is for extracting a representation that is common to all categories from a user feature and an article feature, the category-dependent representation is for determining the categories that the user is interested in from the user feature and the article feature, each piece of the data for training comprises a feature of the user and a feature of an article, the feature of the user and the feature of the article of the piece of the data for training are input into the recommendation model simultaneously, and each piece of the data for training is pre-marked with recommendation information and a category of the article, and the recommendation information indicates whether a feedback on the article is provided by the user; (The specification of data to be stored is understood to be a field of use limitation. See MEPE 2106.05(h)).
...by using a discriminator ...processed by the discriminator... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
...by using a first mapping model ....by using a second mapping model... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
training the recommendation model and the discriminator according to training targets comprising the category-independent representation not corresponding to any one of the plurality of categories, the category-dependent representation corresponding to the pre-marked category, and the prediction result matching with the pre-marked recommendation information, wherein the recommendation model is configured to determine whether to recommend an alternative article to a target user,... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
...and data corresponding to determined articles recommended for the target user are sent to a terminal device of the target user for display (This is understood to be insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
The additional elements as disclosed above alone or in combination do not integrate the judicial exception into practical application as they are mere insignificant extra solution activity in combination of generic computer functions being implemented with generic computer elements in a high level of generality to perform the disclosed abstract idea above.
2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional elements:
A training method of a recommendation model, comprising: (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
...by using a recommendation model... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
inputting the data for training into the recommendation model to obtain an output representation, wherein each piece of the output representation comprises the category-independent representation and the category-dependent representation,... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
...the category-independent representation is for extracting a representation that is common to all categories from a user feature and an article feature, the category-dependent representation is for determining the categories that the user is interested in from the user feature and the article feature, each piece of the data for training comprises a feature of the user and a feature of an article, the feature of the user and the feature of the article of the piece of the data for training are input into the recommendation model simultaneously, and each piece of the data for training is pre-marked with recommendation information and a category of the article, and the recommendation information indicates whether a feedback on the article is provided by the user; (The specification of data to be stored is understood to be a field of use limitation. See MEPE 2106.05(h)).
...by using a discriminator ...processed by the discriminator... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
...by using a first mapping model ....by using a second mapping model... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
training the recommendation model and the discriminator according to training targets comprising the category-independent representation not corresponding to any one of the plurality of categories, the category-dependent representation corresponding to the pre-marked category, and the prediction result matching with the pre-marked recommendation information, wherein the recommendation model is configured to determine whether to recommend an alternative article to a target user,... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
...and data corresponding to determined articles recommended for the target user are sent to a terminal device of the target user for display ( This is directed to well understood, routine of presenting offers and gathering statistics. See MPEP 2106.05 (d)(II)).
The additional elements as disclosed above in combination of the abstract idea are not sufficient to amount to significantly more than the judicial exception as they are mere insignificant extra solution activity in combination of generic computer functions being implemented with generic computer elements in a high level of generality to perform the disclosed abstract idea above.
Regarding claim 9: 2A Prong 1:
processing data to be predicted ...to obtain a category- independent representation and a category-dependent representation,... (mental process – of processing data for training to obtain category- independent representation and a category-dependent representation can be performed by the human mind with the help of pen and paper (e.g., evaluation and judgment)).
determining a prediction result of the data to be predicted according to the data to be predicted, comprising: processing the category-independent representation ...to obtain a first prediction result, or processing the category-independent representation and the category-dependent representation ...to obtain a second prediction result; (mental process – of determining a prediction result according to at least one of the category-independent representation or the category-dependent representation can be performed by the human mind with the help of pen and paper. For example, a person could manually evaluate a category-independent representation and then process the category-independent representation to obtain a first prediction result or manually evaluate category-independent representation and a category-dependent representation, and then process the representations to obtain a second prediction result (e.g., evaluation and judgment)).
determining whether to recommend the alternative article to the target user according to the prediction result of the data to be predicted; and (mental process – of determining whether to recommend the alternative article to the target user according to the prediction result. For example, a person could manually evaluate a prediction result and then determine whether to recommend an alternative article to a target user according to the prediction result (e.g., evaluation and judgment)).
2A Prong 2: This judicial exception is not integrated into a practical application.
Additional elements:
...by using a recommendation model... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
...comprising: inputting the data to be predicted into the recommendation model to obtain an output representation, wherein each piece of the output representation comprises the category-independent representation and the category-dependent representation,... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
...the category-independent representation is for extracting a representation that is common to all categories from a user feature and the article feature, the category-dependent representation is for determining the categories that the user is interested in from the user feature and the article feature, and each piece of the data to be predicted comprises a feature of a target user and a feature of an alternative article, and the feature of the user and the feature of the article of the piece of the data for training are input into the recommendation model simultaneously; (The specification of data to be stored is understood to be a field of use limitation. See MEPE 2106.05(h)).
...by using a first mapping model ...by using a second mapping model... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
sending data corresponding to determined articles recommended for the target user to a terminal device of the target user for display (This is understood to be insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
The additional elements as disclosed above alone or in combination do not integrate the judicial exception into practical application as they are mere insignificant extra solution activity in combination of generic computer functions being implemented with generic computer elements in a high level of generality to perform the disclosed abstract idea above.
2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional elements:
...by using a recommendation model... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
...comprising: inputting the data to be predicted into the recommendation model to obtain an output representation, wherein each piece of the output representation comprises the category-independent representation and the category-dependent representation,... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
...the category-independent representation is for extracting a representation that is common to all categories from a user feature and the article feature, the category-dependent representation is for determining the categories that the user is interested in from the user feature and the article feature, and each piece of the data to be predicted comprises a feature of a target user and a feature of an alternative article, and the feature of the user and the feature of the article of the piece of the data for training are input into the recommendation model simultaneously; (The specification of data to be stored is understood to be a field of use limitation. See MEPE 2106.05(h)).
...by using a first mapping model ...by using a second mapping model... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
sending data corresponding to determined articles recommended for the target user to a terminal device of the target user for display ( This is directed to well understood, routine of receiving or transmitting data over a network. See MPEP 2106.05 (d)(II)).
The additional elements as disclosed above in combination of the abstract idea are not sufficient to amount to significantly more than the judicial exception as they are mere insignificant extra solution activity in combination of generic computer functions being implemented with generic computer elements in a high level of generality to perform the disclosed abstract idea above.
Regarding claim 12: is rejected under the same rational of claim 1. Claim 12 only recites the additional elements of A training apparatus of a recommendation model, comprising: a memory; and a processor coupled to the memory, the processor configured to, based on instructions stored in the memory, perform a training method comprising... which is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f).
Regarding claim 2: Depends on claim 1, thus the rejection of claim 1 is incorporated.2A Prong 1:
wherein the discrimination result ...has a plurality of dimensions corresponding one-to-one to the plurality of categories, and a value of each dimension of the dimensions represents a probability that a representation processed by the discriminator is related to a category corresponding to the dimension (mental process – of evaluation and judgment that can be performed by the human mind with the help of pen and paper. For example, a person could manually evaluate the representations and then, when processing the representations to obtain discrimination results, could determine to include a plurality of dimensions corresponding one-to-one to the plurality of categories and that a value of each dimension represents a probability that a representation is related to a category corresponding to the dimension).
2A Prong 2 and 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional elements:
...of the discriminator... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
...by the discriminator... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
Regarding claim 3: Depends on claim 2, thus the rejection of claim 2 is incorporated.2A Prong 1:
wherein the training the recommendation model and the discriminator comprises: determining a first loss value according to the discrimination results corresponding to the category-independent representation from the discriminator and a category-independent target result, wherein a value of each dimension in the category-independent target result is lower than a low threshold; and (mental process – of evaluation and judgment that can be performed by the human mind with the help of pen and paper. For example, a person could manually evaluate discrimination results and a target result, and then determine a first loss value according to the discrimination results and the target result).
adjusting parameters of the recommendation model and parameters of the discriminator based on the first loss value (mental process – of evaluation and judgment that can be performed by the human mind with the help of pen and paper. For example, a person could manually evaluate parameters of a recommendation model, parameters of a discriminator, and a first loss value, and then adjust the parameters based on the first loss value).
2A Prong 2 and 2B: None.
Regarding claim 4: Depends on claim 3, thus the rejection of claim 3 is incorporated.2A Prong 1:
wherein: the training the recommendation model and the discriminator further comprises: determining a second loss value according to the first prediction result and the pre-marked recommendation information to adjust the parameters of the recommendation model, the parameters of the discriminator and parameters of the first mapping model based on the first loss value and the second loss value (mental process – of evaluation and judgment can be performed by the human mind with the help of pen and paper. For example, a person could manually evaluate a first prediction result and pre-marked recommendation information, and then determine a second loss value according to the first prediction result and the pre-marked recommendation information).
2A Prong 2 and 2B: None.
Regarding claim 5: Depends on claim 2, thus the rejection of claim 2 is incorporated.2A Prong 1:
wherein the training the recommendation model and the discriminator comprises:
determining a third loss value according to the discrimination result of the category- dependent representation from the discriminator and a category-dependent target result... (mental process – of evaluation and judgment can be performed by the human mind with the help of pen and paper. For example, a person could manually evaluate a discrimination result and a target result, and then determine a loss value according to the discrimination result and target result).
adjusting parameters of the recommendation model and parameters of the discriminator based on the third loss value (mental process – of evaluation and judgment can be performed by the human mind with the help of pen and paper. For example, a person could manually evaluate the parameters of a recommendation model, the parameters of a discriminator, and a third loss value, and then adjust the parameters based on the third loss value).
2A Prong 2 and 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional elements:
...wherein in the category-dependent target result, a value of a dimension corresponding to a pre-marked category is higher than a high threshold, and values of other dimensions are lower than a low threshold; and (This is directed to restricting the abstract idea to a particular technological environment. See MPEP 2106.05(h)).
Regarding claim 6: Depends on claim 5, thus the rejection of claim 5 is incorporated.2A Prong 1:
Wherein the training the recommendation model and the discriminator further comprises: determining a fourth loss value according to the second prediction result and the pre- marked recommendation information to adjust the parameters of the recommendation model, the parameters of the discriminator and parameters of the second mapping model based on the third loss value and the fourth loss value (mental process – of evaluation and judgment can be performed by the human mind with the help of pen and paper. For example, a person could manually evaluate a second prediction result, and pre-marked recommendation information, and then determine a fourth loss value according to the second prediction result and pre-marked recommendation information).
2A Prong 2 and 2B: None.
Regarding claim 7: Depends on claim 6, thus the rejection of claim 6 is incorporated.2A Prong 1: None.
2A Prong 2 and 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional elements:
wherein in a training process of the recommendation model, the discriminator and the second mapping model, a value of the category-independent representation is set as a specified constant value (This is directed to restricting the abstract idea to a particular technological environment. See MPEP 2106.05(h)).
Regarding claim 8: Depends on claim 1, thus the rejection of claim 1 is incorporated.2A Prong 1: None.
2A Prong 2 and 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional elements:
wherein the recommendation information indicates whether feedback on the article is given by the user (This is directed to restricting the abstract idea to a particular technological environment. See MPEP 2106.05(h)).
Regarding claim 10: Depends on claim 9, thus the rejection of claim 9 is incorporated.2A Prong 1:
...and the determining whether to recommend the alternative article to the target user comprises: determining a reference ranking of the prediction result of the data to be predicted in prediction results corresponding to all articles in the alternative article set; and (mental process – of evaluation and judgment that can be performed by the human mind with the help of pen and paper. For example, a person could manually evaluate prediction results of articles and then determine a reference ranking of the prediction of an article to be predicted in prediction results corresponding to all articles in an article set).
recommending the alternative article to the target user in response to the reference ranking being higher than a predefined ranking (mental process – of evaluation and judgment that can be performed by the human mind with the help of pen and paper. For example, a person could manually evaluate a reference ranking and a predefined ranking, and then recommend an alternative article to a target user if the reference ranking is higher than the predefined ranking).
2A Prong 2 and 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional elements:
wherein the alternative article is in an alternative article set... (This is directed to restricting the abstract idea to a particular technological environment. See MPEP 2106.05(h)).
Regarding Claims 13-16 comprise limitations similar to those of claims 2-5, respectively, and are therefore rejected for at least the same rationale.
Regarding claim 17: Depends on claim 9, thus the rejection of claim 9 is incorporated.2A Prong 1: None.
2A Prong 2 and 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional elements:
An article recommending apparatus, comprising: a memory; and a processor coupled to the memory, the processor configured to, based on instructions stored in the memory, perform the article recommendation method... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
Regarding claim 18: Depends on claim 12, thus the rejection of claim 12 is incorporated.2A Prong 1:
processing data to be predicted ...to obtain a category-independent representation and a category-dependent representation, wherein the data to be predicted comprises a feature of a target user and a feature of an alternative article; (mental process – of evaluation and judgement the can be performed by the human mind with the help of pen and paper. For example, a person could manually evaluate a feature of a target user and a feature of an alternative article, and then process the features to obtain a category-independent and a category-dependent representation).
determining a prediction result of the data to be predicted according to the data to be predicted; (mental process – of evaluation and judgement the can be performed by the human mind with the help of pen and paper. For example, a person could manually evaluate data to be predicted and then determine a prediction result according to the data to be predicted).
determining whether to recommend the alternative article to the target user according to the prediction result of the data to be predicted; and (mental process – of evaluation and judgement the can be performed by the human mind with the help of pen and paper. For example, a person could manually evaluate a prediction result and then determined whether to recommend an alternative article to a target user according to the prediction result).
2A Prong 2 and 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional elements:
An article recommendation system, comprising: the training apparatus of the recommendation model of claim 12; and, an article recommendation apparatus comprising: a memory; and a processor coupled to the memory, the processor configured to, based on instructions stored in the memory, perform an article recommendation method comprising: (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
...by using a recommendation model... (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
sending data corresponding to determined articles recommended for the target user to a terminal device of the target user for display (This is understood to be insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g). Further, this is directed to well understood, routine of presenting offers and gathering statistics. See MPEP 2106.05 (d)(II)).
The additional elements as disclosed above alone or in combination of the abstract idea are not sufficient to amount to significantly more than the judicial exception as they are mere insignificant extra solution activity in combination of generic computer functions being implemented with generic computer elements in a high level of generality to perform the disclosed abstract idea above.
Regarding claim 19: Depends on claim 1, thus the rejection of claim 1 is incorporated.2A Prong 1: None.
2A Prong 2 and 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional elements:
A non-transitory computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the training method according to claim 1 (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
Regarding claim 20: Depends on claim 9, thus the rejection of claim 9 is incorporated.2A Prong 1: None.
2A Prong 2 and 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional elements:
A non-transitory computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the article recommendation method according to claim 9 (This is directed to using computers or other machinery merely as a tool to perform an existing process. See MPEP 2106.05(f)).
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-2, 5-8, 12-13, 16, 18-19 are rejected under 35 U.S.C. 103 as being unpatentable over Chen et al. US 20220374761 A1 (hereinafter Chen) in view of Zhang et al. Content-Collaborative Disentanglement Representation Learning for Enhanced Recommendation (hereinafter Zhang) in further view of Wang eta al. CN 110321952 A (hereinafter Wang) in further view of Courville et al. US 20220036189 A1 (hereinafter Courville) and in further view of Sasaya et al. US 20210232947 A1 (hereinafter Sasaya) in further view of
Regarding claim 1:
Chen teaches A training method of a recommendation model, comprising: processing data for training by using a recommendation model to obtain a category- independent representation and a category-dependent representation, comprising: inputting the data for training into the recommendation model to obtain an output representation, wherein each piece of the output representation comprises the category-independent representation and the category-dependent representation ...each piece of the data for training comprises a feature of the user and a feature of an article,... and each piece of the data for training is pre-marked with recommendation information and a category of the article, and the recommendation information indicates whether a feedback on the article is provided by the user; (Chen [0038] teaches using an embedding model (i.e., recommendation model) to generate a content embedding vector that is calculated by averaging an article word vector (i.e., a category-independent representation); [0040], [0042], and [0048] teach that the embedding model (i.e., recommendation model) may also be used for generating user interest embedding profiles using features of content items, such as topics of news items, with which a user has interacted or skipped (i.e., a category-dependent representation), by inputting a new user click at time t. Furthermore, Chen [0040], [0042], and [0048] teach that the embedding profile is a vector representation of a user profile (i.e., a feature of a user), that is built using the features of the content items that a user has interacted with such as news item topic (i.e., a feature of an article), and a given feature value in the user profile indicates the relevance of the content feature to the user (i.e., pre-marked with recommendation information); [0042] further teaches that the historical user data may be used to construct the training data for that user and that time-decay factors may be taken into account so that the system can keep track of changes in the users’ interests, as changes in the feature (topic) values in the user profile indicate that the relevance of topics to the user is changing. In addition, Chen also teaches [0071] that the recommendation information indicates whether the article will be clicked on (i.e., whether feedback on the article is given) by the user).
processing the category-independent representation by using a discriminator to obtain a first discrimination result corresponding to the category-independent representation, the first discrimination result indicating a correlation between the category-independent representation processed by the discriminator and a plurality of categories, and processing the category-dependent representation by using the discriminator to obtain a second discrimination result corresponding to the category-dependent representation, the second discrimination result indicating a correlation between the category- dependent representation processed by the discriminator and the plurality of categories,... ( Chen ([0071] teaches that the ranking model (i.e., a discriminator) processes the user click history representation (i.e., which includes features of the clicked articles, such as article category as taught by [0040-0041], and is a category-dependent representation) and the candidate news article representation (i.e., a category-independent representation, as taught by [0038]) to determine a prediction score that represents a probability that a user will click on it or not (i.e., a correlation between the user/article representations and a plurality of categories); [0038] teaches using an embedding model (i.e., recommendation model) to generate the first vector, content embedding, which serves to generate a mean pooling of the words of an article that is calculated by averaging an article word vector (i.e., a category-independent representation) and [0040, 0042, and 0048] teach that the embedding model (i.e., recommendation model) may also be used for generating user interest embedding profiles using features of content items, such as topics of news items, with which a user has interacted or skipped (i.e., a category-dependent representation), by inputting a new user click at time t (i.e., a user feature)) - that is, a feature of a target user and a feature of an alternative article are input to the embedding model (i.e., inputting the data to be predicted into the recommendation model) as a step in generating a representation comprising a category‐independent representation and category‐dependent representation to be output from the representation generation phase (i.e., the two representations are collectively an output representation) and provided to the ranking model (i.e., the discriminator);
determining a prediction result according to at least one of the category-independent representation or the category-dependent representation, comprising: (Examiner will like to emphasize the claim as presented recites determining a prediction result according to at least one of the category-independent representation or the category-dependent representation for which Chen [0071] teaches determining a list of news articles to recommend to the user (i.e., prediction result) based on the ranking model’s prediction score, which was generated by processing the candidate news article (i.e., a category-independent representation as taught by [0038]).
processing the category-independent representation to obtain a first prediction result, or processing the category-independent representation and the category-dependent representation to obtain a second prediction result; (Examiner will like to emphasize the claim as presented recites processing the category-independent representation by using a first mapping model... or processing the category-independent representation and the category-dependent representation by using a second mapping model.. for which Chen [0071] teaches the ranking model process candidate news article (i.e., a category-independent representation as taught by [0038]) to generate a list of news articles (i.e., first prediction result) to recommend to the user based on the ranking model's prediction score).
training and the discriminator according to training targets comprising the category-independent representation not corresponding to any one of the plurality of categories, the category-dependent representation corresponding to the pre-marked category, and the prediction result matching with the pre-marked recommendation information... (In this limitation, essentially an output of the discriminator is being used as the basis for training both the recommendation model and the discriminator model. Chen discloses training an embedding model (i.e., the recommendation model) in [0034-0035], but not explicitly based on the output of the discriminator. [0071] teaches that the ranking model (i.e., the discriminator) is trained using a training set of size N and according to a negative log-likelihood objective function that compares the ranking model output and the actual category label);
...wherein the recommendation model is configured to determine whether to recommend an alternative article to a target user, and data corresponding to determined articles recommended for the target user are sent to a terminal device of the target user for display (Chen [0054], [0061], and Figure 1B teach serving content items according to one implementation of the present disclosure, wherein the one or more content items recommended by the content recommendation system (i.e., including the recommendation model) are transmitted over a network to the client device for rendering thereon).
Chen does not explicitly teach: ...the category-independent representation is for extracting a representation that is common to all categories from a user feature and an article feature, the category-dependent representation is for determining categories that the user is interested in from the user feature and the article feature, ...the feature of the user and the feature of the article of the piece of the data for training are input into the recommendation model simultaneously; ...wherein the processing the category-independent representation and the processing the category-dependent representation are performed separately; determining a prediction result of the data to be predicted according to the data to be predicted, comprising: processing the category-independent representation by using a first mapping model to obtain a first prediction result, or processing the category-independent representation and the category-dependent representation by using a second mapping model to obtain a second prediction result; and training the recommendation model and discriminator… according to training targets ...and the prediction result matching with … information
However, Zhang teaches the following:
...the category-independent representation is for extracting a representation that is common to all categories from a user feature and an article feature, the category-dependent representation is for determining categories that the user is interested in from the user feature and the article feature, each piece of the data for training comprises a feature of the user and a feature of an article, the feature of the user and the feature of the article of the piece of the data for training are input into the recommendation model simultaneously,... ( Zhang, Abstract propose a novel two-level disentanglement generative recommendation model (DICER) that supports both content collaborative disentanglement and feature disentanglement. Specifically, pg. 3, sec. Method, para 1 teaches to learn user preference, two type of information are used as input for the recommendation model respectively that is the user-item interactions capturing the implicit feedback from user towards item and item content information for each item, that “could correspond to item images, reviews, descriptive text or other item specific information. In addition, pg. 3, sec. Method, para 2, & FIG. 1(b) teaches how DICER can be used to disentangle a user preference over an item in order to obtain content features
z
i
C
(category-independent representation) extracted from item content
c
i