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 .
Status of Claims
This action is in reply to the application filed on 07/01/2025.
Claims 1-19 are currently pending and have been examined.
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 1-19 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
In the instant case, claims 1, 10 and 15 are directed to a method, system, and non-transitory computer-readable recording medium.
For the purposes of this analysis, representative claim 1 is addressed. (Step 2A, prong 1) Abstract ideas are in bold below, and represents a “predicting future returns of a financial market” which is a grouped under “Certain methods of organizing human activity — fundamental economic practices” in prong one of step 2A (MPEP 2106.04(a)).
A method, comprising:
training a machine-learning model with financial data;
determining a portion of the financial data to input to the trained model;
predicting, with the machine-learning model in response to the portion of the financial data, future returns of a financial market during a time window; and
constructing an investment portfolio of one or more assets of the financial market in response to the future returns.
The additional elements of claim 1 such as “training a machine-learning model with financial data”, “determining a portion of the financial data to input to the trained model”, “…with the machine-learning model in response to the portion of the financial data…”, represent the use of a computer as a tool to perform an abstract idea and/or does no more than generally link the abstract idea to a particular field of use. Independent claims 10 and 15 recite similar additional elements as independent claim 1 and further additional elements of “an electronic computing circuit, and a tangible, non-transitory computer-readable medium storing instructions executed by a computing circuit or another electric circuit” which also represent the use of a computer as a tool to perform an abstract idea and/or does no more than generally link the abstract idea to a particular field of use
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 integration into a practical application, the additional elements amount to no more than mere instructions to apply the abstract idea of using generic computer components. The claim elements when considered separately and in an ordered combination, do not add significantly more than implementing the abstract idea of “predicting future returns of a financial market”
Hence, claims 1, 10 and 15 are not patent eligible.
Dependent claims 2-7, 9-14, and 16-20 recited additional details which only further narrow the abstract idea and do not add any additional features, alone or in combination, that would provide a practical application or provide significantly more.
Claims 2 and 11 recites additional elements of “wherein training the machine-learning model comprises training the machine-learning model with a walk-forward cross-validation strategy.” does no more than use a computer as a tool to perform an abstract idea and do no more than generally link the abstract idea to a particular field of use. Therefore, as it is no more than apply it does not improve the functioning of a computer, or improve other technology or technical field.
Claim 3 and 12 recites the additional elements of “wherein training the machine-learning model comprises training and cross-validating the machine-learning model with a number of folds.” does no more than use a computer as a tool to perform an abstract idea and do no more than generally link the abstract idea to a particular field of use. Therefore, as it is no more than apply it does not improve the functioning of a computer, or improve other technology or technical field.
Claims 4 and 13 recite the additional elements of “wherein determining a portion of the financial data to input to the trained model comprises iteratively removing least-important data features from the portion of the financial data until only a threshold number of data features are left.” does no more than use a computer as a tool to perform an abstract idea and do no more than generally link the abstract idea to a particular field of use. Therefore, as it is no more than apply it does not improve the functioning of a computer, or improve other technology or technical field.
Claim 5 and 14 recites the additional elements of “wherein training the machine-learning model and determining a portion of the financial data to input to the trained model comprises: training the model with the financial data; assigning, with the model, a respective importance score to each feature; removing at least one feature in response to the respective importance score of each of the at least one feature; and repeating the training, assigning, and removing at least one time.” does no more than use a computer as a tool to perform an abstract idea and do no more than generally link the abstract idea to a particular field of use. Therefore, as it is no more than apply it does not improve the functioning of a computer, or improve other technology or technical field.
Claims 6 and 16 recite the additional elements of “wherein training the machine-learning model and determining a portion of the financial data to input to the trained model comprises: training the model with the financial data; assigning, with the model, a respective importance score to each feature; removing at least one feature having a lowest importance score; and repeating the training, assigning, and removing at least one time.” does no more than use a computer as a tool to perform an abstract idea and do no more than generally link the abstract idea to a particular field of use. Therefore, as it is no more than apply it does not improve the functioning of a computer, or improve other technology or technical field.
Claims 7 and 17 recite the additional elements of “generating the financial data by: parsing and flattening a first set of financial data; combining the parsed-and-flattened first set of financial data with a second set of financial data.” does no more than use a computer as a tool to perform an abstract idea and do no more than generally link the abstract idea to a particular field of use. Therefore, as it is no more than apply it does not improve the functioning of a computer, or improve other technology or technical field.
Claims 8 and 18 recite the additional elements of “before predicting, with the machine-learning model, the future returns of the financial market, evaluating the model using the Spearman rank correlation coefficient.” does no more than use a computer as a tool to perform an abstract idea and do no more than generally link the abstract idea to a particular field of use. Therefore, as it is no more than apply it does not improve the functioning of a computer, or improve other technology or technical field.
Claims 9 and 19 recite the additional elements of “wherein constructing the investment portfolio comprises estimating risk of the investment portfolio using a nested clustered optimization methodology.” does no more than use a computer as a tool to perform an abstract idea and do no more than generally link the abstract idea to a particular field of use. Therefore, as it is no more than apply it does not improve the functioning of a computer, or improve other technology or technical field.
The claims as a whole do not amount to significantly more than the abstract idea itself. This is because the claims do not affect an improvement to another technology or technical field, the claims do not amount to an improvement to the functioning of a computer system itself, and the claims do not move beyond a general link of the use of an abstract idea to a particular technological environment.
Accordingly, there are no meaningful limitations in the claims that transform the judicial exception into a patent eligible application such that the claims amount to significantly more than the judicial exception itself.
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.
Claims 1, 5-6, 10, 14-16 are rejected under 35 U.S.C. 103 as being unpatentable over Wu at al. (US 2024/0144372 A1) in view of Carter (US 2024/0087029 A1).
Regarding claims 1, 10 and 15
training a machine-learning model with financial data; determining a portion of the financial data to input to the trained model; (See at least Wu [0018] Systems and methods are disclosed related to interactive systems that use neural networks to determine financial investment predictions and recommendations. For instance, a system(s) may obtain, generate, retrieve, and/or receive data for processing using one or more neural networks that are trained to determine financial predictions associated with financial investments. The data may include, but is not limited to, user data, news data, financial data, prediction data, and/or any other type of data.)
predicting, with the machine-learning model in response to the portion of the financial data, future returns of a financial market during a time window; and (See at least Wu [0019] The system(s) may then input the data into the neural network(s) that processes the data and, based on the processing, determines the financial predictions associated with investments. As described herein, an investment may include, but is not limited to, a stock, a bond, a mutual fund, real estate, an ETF, a cash bank deposit, a cryptocurrency, a commodity, and/or other type of investment. Additionally, a financial prediction for an investment may include, but is not limited to, a predicted movement of the investment (e.g., extremely down, down, preserved, up, extremely up, etc.), a predicted price of the investment (e.g., a future stock price, etc.), a recommendation to acquire, hold, or sell the investment, and/or any other financial prediction or recommendation.)
Wu does not specifically teach: constructing an investment portfolio of one or more assets of the financial market in response to the future returns.
However Carter teaches: [0021] The systems and methods disclosed herein provide various technical solutions within the wealth management space, including but not limited to, (i) automatically provisioning personalized portfolios on demand and in real-time; (ii) incorporating data from disparate sources to generate the personalized portfolios; and (iii) providing a secure, centralized access point for generating a personalized portfolio.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify financial investment predications using neural networks of Wu in view of method for an artificial intelligence enabled processing of personalized autonomous portfolios as taught by Carter in order to use financial predictions associated with financial investments to create portfolios.
Regarding claims 5 and 14
wherein training the machine-learning model and determining a portion of the financial data to input to the trained model comprises: training the model with the financial data; (See at least Wu [0018] Systems and methods are disclosed related to interactive systems that use neural networks to determine financial investment predictions and recommendations. For instance, a system(s) may obtain, generate, retrieve, and/or receive data for processing using one or more neural networks that are trained to determine financial predictions associated with financial investments. The data may include, but is not limited to, user data, news data, financial data, prediction data, and/or any other type of data.)
assigning, with the model, a respective importance score to each feature; (See at least Wu [0109] The financial vectors 418 may represent one or more past prices associated with an investment being analyzed by the neural network(s) 416 (e.g., each financial vector 418 may represent one past price, two past prices, five past prices, etc.). In some examples, the financial vectors 418 may represent the prices using time intervals, such as every fifteen minutes, every thirty minutes, every hour, every day, every week, every month, every year, and/or any other time interval. For example, if the time interval is fifteen minutes, then a first financial vector 418 may represent a first price of the investment at a current time, a second financial vector 418 may represent a second price of the investment fifteen minutes before the current time, a third financial vector 418 may represent a third price of the investment thirty minutes before the current time, a fourth financial vector 418 may represent a fourth price of the investment forty-five minutes before the current time, a fifth financial vector 418 may represent a fifth price of the investment an hour before the current time, and/or so forth. In some examples, the financial vectors 418 are associated with a first time period. The first time period may include, but is not limited to, one day, one week, one month, one year, and/or any other period of time.)
removing at least one feature in response to the respective importance score of each of the at least one feature; and (See at least Wu [0110] For example, if the time interval is fifteen minutes and the first time period is one hour, then there may be five financial vectors 418 input into neural network(s) 416 (the first financial vector 418 through the fifth financial vector 418 from the example above). In such an example, the financial vectors 418 may be updated for processing using the neural network(s) 416 based on the time interval. For instance, and using the example above where the time interval is fifteen minutes and the first time period is one hour, after an elapse of fifteen minutes, the fifth financial vector 418 (e.g., the vector representing the oldest price of the investment) may be removed from the inputs and a sixth financial vector 418 representing the new current price of the investment may be generated and used as an input. In other words, the inputs to the neural network(s) 416 may include a rolling window of the financial vectors 418.)
repeating the training, assigning, and removing at least one time. (See at least Wu [0110] In other words, the inputs to the neural network(s) 416 may include a rolling window of the financial vectors 418.)
Regarding claims 6 and 16
wherein training the machine-learning model and determining a portion of the financial data to input to the trained model comprises: training the model with the financial data; (See at least Wu [0018] Systems and methods are disclosed related to interactive systems that use neural networks to determine financial investment predictions and recommendations. For instance, a system(s) may obtain, generate, retrieve, and/or receive data for processing using one or more neural networks that are trained to determine financial predictions associated with financial investments. The data may include, but is not limited to, user data, news data, financial data, prediction data, and/or any other type of data.)
assigning, with the model, a respective importance score to each feature; (See at least Wu [0109] The financial vectors 418 may represent one or more past prices associated with an investment being analyzed by the neural network(s) 416 (e.g., each financial vector 418 may represent one past price, two past prices, five past prices, etc.). In some examples, the financial vectors 418 may represent the prices using time intervals, such as every fifteen minutes, every thirty minutes, every hour, every day, every week, every month, every year, and/or any other time interval. For example, if the time interval is fifteen minutes, then a first financial vector 418 may represent a first price of the investment at a current time, a second financial vector 418 may represent a second price of the investment fifteen minutes before the current time, a third financial vector 418 may represent a third price of the investment thirty minutes before the current time, a fourth financial vector 418 may represent a fourth price of the investment forty-five minutes before the current time, a fifth financial vector 418 may represent a fifth price of the investment an hour before the current time, and/or so forth. In some examples, the financial vectors 418 are associated with a first time period. The first time period may include, but is not limited to, one day, one week, one month, one year, and/or any other period of time.)
removing at least one feature having a lowest importance score; and (See at least Wu [0110] For example, if the time interval is fifteen minutes and the first time period is one hour, then there may be five financial vectors 418 input into neural network(s) 416 (the first financial vector 418 through the fifth financial vector 418 from the example above). In such an example, the financial vectors 418 may be updated for processing using the neural network(s) 416 based on the time interval. For instance, and using the example above where the time interval is fifteen minutes and the first time period is one hour, after an elapse of fifteen minutes, the fifth financial vector 418 (e.g., the vector representing the oldest price of the investment) may be removed from the inputs and a sixth financial vector 418 representing the new current price of the investment may be generated and used as an input. In other words, the inputs to the neural network(s) 416 may include a rolling window of the financial vectors 418.)
repeating the training, assigning, and removing at least one time.
[0110] For example, if the time interval is fifteen minutes and the first time period is one hour, then there may be five financial vectors 418 input into neural network(s) 416 (the first financial vector 418 through the fifth financial vector 418 from the example above). In such an example, the financial vectors 418 may be updated for processing using the neural network(s) 416 based on the time interval. For instance, and using the example above where the time interval is fifteen minutes and the first time period is one hour, after an elapse of fifteen minutes, the fifth financial vector 418 (e.g., the vector representing the oldest price of the investment) may be removed from the inputs and a sixth financial vector 418 representing the new current price of the investment may be generated and used as an input. In other words, the inputs to the neural network(s) 416 may include a rolling window of the financial vectors 418.
Claims 2 and 11 are rejected under 35 U.S.C. 103 as being unpatentable over Wu at al. (US 2024/0144372 A1) in view of Carter (US 2024/0087029 A1) and further in view of Bean et al. (US 2022/0353273 A1).
Regarding claims 2 and 11
Wu does not specifically teach: wherein training the machine-learning model comprises training the machine-learning model with a walk-forward cross-validation strategy.
However Bean teaches: [0056] Further, improvements may be made to the machine-learning model by including a model that is less prone to overfitting, by adjusting hyperparameters of the model, or by increasing training data provided to the model during training operations. Moreover, a walk-forward cross-validation method may be implemented to validate the machine-learning model during training. Implementing such a technique may provide critical information on how to select the most appropriate retraining cadence for the machine-learning model while also providing a framework for hyperparameter tuning of the model. In some examples, the prediction and confidence threshold for the model are optimized using tools such as the Spark machine learning library (MLLib) to optimize for the recall of the model.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify financial investment predications using neural networks of Wu in view of machine learning techniques for object authentications as taught by Bean in order to use walk forward cross-validation method for validating machine learning training.
Claims 3 and 12 are rejected under 35 U.S.C. 103 as being unpatentable over Wu at al. (US 2024/0144372 A1) in view of Carter (US 2024/0087029 A1) and further in view of O'Rourke (US 2012/0226645 A1).
Regarding claims 3 and 12
Wu does not specifically teach: wherein training the machine-learning model comprises training and cross-validating the machine-learning model with a number of folds.
However O'Rourke teaches: [0044] In an exemplary embodiment, the combining function of the prediction software application includes a smoothing function using K-fold cross-validation and bagging (or bootstrap aggregation), with multiple sub-samples from the received data values to identify robust influential features and to prevent over-fitting during the training phase. Exemplary smoothing functions that may be applied include loess and kernel smoothers and radial basis function.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify financial investment predications using neural networks of Wu in view of the predicting the performance of a financial instrument as taught by O’Rourke in order to use machine learning to predict future valuation of financial instruments.
Claims 4 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Wu at al. (US 2024/0144372 A1) in view of Carter (US 2024/0087029 A1) and further in view of Sinha Roy et al. (2024/0330779 A1)
Regarding claims 4 and 13
Wu does not specifically teach: wherein determining a portion of the financial data to input to the trained model comprises iteratively removing least-important data features from the portion of the financial data until only a threshold number of data features are left.
However Sinha Roy teaches: [0086] The system iteratively trains random forest models, with each iteration removing the least-important feature. A user or system may specify a number of features to be included in the final trained model 315
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify financial investment predications using neural networks of Wu in view of the machine learning model based life cycle classification for selection furcate model as taught by Sinha Roy in order to improve forecasting machine learning models.
Claims 7 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Wu at al. (US 2024/0144372 A1) in view of Carter (US 2024/0087029 A1) and further in view of De Shetler (US 2021/0065191 A1)
Regarding claims 7 and 17
Wu does not specifically teach: parsing and flattening a first set of financial data; combining the parsed-and-flattened first set of financial data with a second set of financial data.
However De Shetler teaches: [0070] Training the machine learning model may be an optional step, not shown in FIG. 5A. The machine learning model may be trained using data describing past merchant behavior by many merchants over a time period, such as 90 days. The training data set includes a description of each merchant, each merchant's credit score, and instances where merchants failed to satisfy their reimbursement obligation for chargebacks under the terms of the TPEP system use agreement. Merchants' financial information may also be available via a financial management application which is used by the merchants to track their financial affairs and taxes. After the data is obtained, the data is parsed and flattened as described above, to establish known data in a machine readable vector format suitable for use by the machine learning model. Part of the known data is held back. The remaining portion of the data is provided to the machine learning model, and the machine learning model is instructed to identify patterns in the features that correlate with a merchant failing to meet a financial obligation to reimburse a chargeback. Later, the held back data is provided to the machine learning model, but this time the data describing which merchants failed to reimburse for chargebacks are also held back. The machine learning model then predicts the probability that the identified patterns of features will correspond to a merchant that will fail to reimburse a chargeback.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify financial investment predications using neural networks of Wu in view of the Machine learning based determination of limits on merchant use of a third party payment system as taught by Sinha Roy in order to improve training of machine learning models.
Claims 8 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Wu at al. (US 2024/0144372 A1) in view of Carter (US 2024/0087029 A1) and further in view of Culter (US 2023/0125150 A1)
Regarding claims 8 and 18
Wu does not specifically teach: before predicting, with the machine-learning model, the future returns of the financial market, evaluating the model using the Spearman rank correlation coefficient.
However Culter teaches: [0077] One evaluation task in noise suppression is to compare the speech quality produced by different models on the same audio file. Given N noise suppressors, step 4 of the above pipeline can be adjusted by sampling within each cluster audio file with a probability proportional to the variance across models of the predicted dMOS for a given audio file. Instead of the sampling error (1), sampling performance can be measured with the Spearman’s rank correlation coefficient between the ranking of the N models obtained on the sample S and the ranking obtained with the entire target data D.sub.t.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify financial investment predications using neural networks of Wu in view of the augmentation of testing or training sets for machine learning models as taught by Culter in order to improve training of machine learning models.
Claims 9 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Wu at al. (US 2024/0144372 A1) in view of Carter (US 2024/0087029 A1) and further in view of (Lopez de Prado (US 2021/0081828 A1)
Regarding claims 9 and 19
Wu does not specifically teach: wherein constructing the investment portfolio comprises estimating risk of the investment portfolio using a nested clustered optimization methodology.
However Lopez de Prado teaches: [0077] [0060] According to some embodiments of the present invention, there are provided methods and systems for enhancing the convex optimization based prediction models to significantly reduce their computation load while increasing their performance and robustness. In particular, the computation load is reduced while the performance and robustness are increased by applying a Nested Clustered Optimization (NCO) for the convex optimization based prediction models.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify financial investment predications using neural networks of Wu in view of the applying monto carol and machine learning methods for robust conveys optimization based prediction algorithms as taught by Lopez de Prado in order to improve performance and robustness of prediction models.
Prior Art of Record Not Currently Relied Upon
Huang et al (US 2019/0180375 A1) Teaches: Financial risk forecast method
Rebuth et al (US 2022/058738 A1) Teaches: Method for dynamic asset portfolio allocation and management.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to GREGORY MARK JAMES whose telephone number is (571)272-5155. The examiner can normally be reached M-F 8:30am - 5:00pm EST.
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, Ryan Donlon can be reached at 571-270-3602. 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.
/GREGORY M JAMES/Examiner, Art Unit 3692
/RYAN D DONLON/Supervisory Patent Examiner, Art Unit 3692 August 4, 2026
Prosecution Insights