DETAILED ACTION
Status of Claims
1. This is a Final office action in response to communication received on 07/23/2023. Claims 1-18 are pending and examined herein.
Priority
2. The examiner acknowledges priority benefits being claimed by the Applicant for U.S. Provisional Application No. 63/313,609 filed on February 24, 2022. However the claims as supported by Figs. 7-9 and their associated disclosure are not entitled to the priority benefit as that disclosure was filed in the Non-provisional 18/172,634 filed on 02/22/2023.
Claim Rejections - 35 USC § 101
3. 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 1-18 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. Next using the 2019 Revised Patent Subject Matter Eligibility Guidances (hereinafter 2019 PEG) the rejection as follows has been applied.
Under step 1, analysis is based on MPEP 2106.03, Claims 1-8 are a non-transitory computer readable medium; and claims 9-18 are a method. Thus, each claim 1-18, on its face, is directed to one of the statutory categories (i.e., useful process, machine, manufacture, or composition of matter) of 35 U.S.C. §101.
Under Step 2A Prong One, per MPEP 2106.04, prong one asks does the claim recite an abstract idea, law of nature, or natural phenomenon? In Prong One examiners evaluate whether the claim recites a judicial exception, i.e. whether a law of nature, natural phenomenon, or abstract idea is set forth or described in the claim. While the terms "set forth" and "described" are thus both equated with "recite", their different language is intended to indicate that there are two ways in which an exception can be recited in a claim. For instance, the claims in Diehr, 450 U.S. at 178 n. 2, 179 n.5, 191-92, 209 USPQ at 4-5 (1981), clearly stated a mathematical equation in the repetitively calculating step, and the claims in Mayo, 566 U.S. 66, 75-77, 101 USPQ2d 1961, 1967-68 (2012), clearly stated laws of nature in the wherein clause, such that the claims "set forth" an identifiable judicial exception. Alternatively, the claims in Alice Corp., 573 U.S. at 218, 110 USPQ2d at 1982, described the concept of intermediated settlement without ever explicitly using the words "intermediated" or "settlement."
Next, per 2019 PEG, to determine whether a claim recites an abstract idea in Prong One, examiners are now to: (I) Identify the specific limitation(s) in the claim under examination (individually or in combination) that the examiner believes recites an abstract idea; and (II) determine whether the identified limitation(s) falls within the subject matter groupings of abstract ideas enumerated in Section I of the 2019 PEG. If the identified limitation(s) falls within the subject matter groupings of abstract ideas enumerated in Section I, analysis should proceed to Prong Two in order to evaluate whether the claim integrates the abstract idea into a practical application.
(I) An abstract idea as recited per abstract recitation of claims 1-8, and 9-18 [i.e. recitation with the exception of additional elements as noted and analyzed under step 2A prong two and step 2B inquiries below, i.e. under step 2A prong one the Examiner considered claim recitation other than the additional elements (which once again are expressly noted below) to be the abstract recitation] (II) is that of predicting performance using a model during a target time period, by training said model on past or historic or a time preceding the target time period, and comparing the predicted performance during the target time period without advertising campaign with performance of the company with the advertising campaign during the target time period to indicate difference in performance or to evaluate impact of the advertising campaign which is certain methods of organizing human activity.
The phrase "Certain methods of organizing human activity" applies to fundamental economic principles or practices (including hedging, insurance, mitigating risk); commercial or legal interactions (including agreements in the form of contracts; legal obligations; advertising, marketing or sales activities or behaviors; business relations)); managing personal behavior or relationships or interactions between people (including social activities, teaching, and following rules or instructions). Further, see MPEP 2106.04(a)(2) II. A-C.
Therefore, the identified limitations fall within the subject matter groupings of abstract ideas enumerated in Section I of 2019 PEG, thus analysis now proceeds to Prong Two in order to evaluate whether the claim integrates the abstract idea into a practical application.
Under Step 2A Prong Two, per MPEP 2106.04, prong two asks does the claim recite additional elements that integrate the judicial exception into a practical application? In Prong Two, examiners evaluate whether the claim as a whole integrates the exception into a practical application of that exception. If the additional elements in the claim integrate the recited exception into a practical application of the exception, then the claim is not directed to the judicial exception (Step 2A: NO) and thus is eligible at Pathway B. This concludes the eligibility analysis. If, however, the additional elements do not integrate the exception into a practical application, then the claim is directed to the recited judicial exception (Step 2A: YES), and requires further analysis under Step 2B (where it may still be eligible if it amounts to an ‘‘inventive concept’’).
Next, per 2019 PEG, Prong Two represents a change from prior guidance. The analysis under Prong Two is the same for all claims reciting a judicial exception, whether the exception is an abstract idea, a law of nature, or a natural phenomenon. Examiners evaluate integration into a practical application by: (I) Identifying whether there are any additional elements recited in the claim beyond the judicial exception(s); and (II) evaluating those additional elements individually and in combination to determine whether they integrate the exception into a practical application, using one or more of the considerations laid out by the Supreme Court and the Federal Circuit.
Accordingly, the examiner will evaluate whether the claims recite one or more additional element(s) that integrate the exception into a practical application of that exception by considering them both individually and as a whole.
The claim elements in addition to the abstract idea, i.e. additional elements, as recited in claims 1-18 at least are a non-transitory medium, a processor of a computing device, digital presentation on an interface to visually and programmatically depict data, and machine learning algorithm/model (claim 1); and a computer program executing on a computing device and machine learning algorithm/model (claim 9); selection through an interface (claim 15); REST API or database connector (claim 16); digital presentation of an interface to upload files (claim 17); CSV or spreadsheet files (claim 18). Remaining claims, namely 2-8, and 10-14, either recite the same additional element(s) as already noted above or simply lack recitation of an additional element, in which case note prong one as set forth above.
As would be readily apparent to a person having ordinary skill in the art (hereinafter PHOSITA), the additional elements are generic computer components. The additional elements including machine learning algorithm/model are simply utilized as generic tools to implement the abstract idea or plan as "apply it" instructions (see MPEP 2106.05(f)). The additional elements are generic as they are described at a high level of generality, see at least as-filed Figs. 6-7 and 10 and their associated disclosure and see at least as filed spec. paras. [0013]-[0031] and [0045]-[0050]; and see [0025]-[0028] as it pertains to high level description of machine learning model.
The processor executing the "apply it" instruction is further able to sending/receiving/upload data over a network, note 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); TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016) (using a telephone for image transmission); 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) which is considered insignificant extra solution activity (see MPEP 2106.05(g)). Thus, the process is similar to collecting information, analyzing it, and displaying certain results of the collection and analysis (Electric Power Group).
The abstract idea is intended to be merely carried out in a technical environment such as collecting/communicating data via a network and analyzing data via a generic processor to evaluate impact of marketing campaign during a target time period by comparing it with control (see MPEP 2106.05(h)).
Accordingly, viewed as a whole, these additional claim element(s) do not provide any additional element that integrates the abstract idea (prong one), into a practical application (prong two) upon considering the additional elements both individually and as a combination or as a whole as they fail to provide: an additional element that reflects an improvement in the functioning of a computer, or an improvement to other technology or technical field; or an additional element that implements a judicial exception with, or uses a judicial exception in conjunction with, a particular machine or manufacture that is integral to the claim; or an additional element that effects a transformation or reduction of a particular article to a different state or thing; or an additional element that applies or uses the judicial exception, again, in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment, such that the claim as a whole is more than a drafting effort designed to monopolize the exception as explained above.
Thus, the abstract idea of predicting performance during a target time period and comparing the predicted performance during the target time period without advertising campaign with performance of the company with the advertising campaign during the target time period to indicate difference in performance or to evaluate impact of the advertising campaign which is certain methods of organizing human activity (prong one) is not integrated into a practical application upon consideration of the additional element(s) both individually and as a combination (prong two).
Therefore, under step 2A, the claims are directed to the abstract idea, and require further analysis under Step 2B.
Under step 2B, per MPEP 2106.05, as it applies to claims 1-18, the Examiner will evaluate whether the foregoing additional elements analyzed under prong two, when considered both individually and as a whole provide an inventive concept (i.e., whether the additional elements amount to significantly more than the exception itself). The abstract idea of predicting performance during a target time period and comparing the predicted performance during the target time period without advertising campaign with performance of the company with the advertising campaign during the target time period to indicate difference in performance or to evaluate impact of the advertising campaign which is certain methods of organizing human activity - has not been applied in an eligible manner. The claim elements in addition to the abstract idea are simply being utilized as generic tools to execute "apply it" instructions as they are described at a high level of generality. Additionally, the abstract idea is intended to be merely carried out in a technical environment, however fail to contain meaningful limitations beyond generally linking the use of an abstract idea to a particular technological environment (Id. or note step 2A prong two).
Regarding, insignificant solution activity such as data gathering or post solution activity such as displaying on interface, the Examiner relies on court cases and publications that demonstrate that such a way to gather data and display information is indeed well-understood, routine, or conventional in the industry or art, at least note as follows:
(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); TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016) (using a telephone for image transmission); 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) [similarly here datasets are received]; and
(ii) (a)Affinity v DirecTV - "The court rejected the argument that the computer components recited in the claims constituted an “inventive concept.” It held that the claims added “only generic computer components such as an ‘interface,’ ‘network,’ and ‘database,’” and that “recitation of generic computer limitations does not make an otherwise ineligible claim patent-eligible.” Id. at 1324-25 (citations omitted). The court noted that nothing in the asserted claims purported to improve the functioning of the computer itself or “effect an improvement in any other technology or technical field.” Mortgage Grader, 811 F.3d at 1325 (quoting Alice, 134 S. Ct. at 2359)."; and (b) (ii) Collecting and analyzing information to detect misuse and notifying a user when misuse is detected [similarly here datasets are presented on interface and/or interface is provided for dataset selection; and also based on abstract evaluation/comparison of plotted data sets for with and without marketing campaign a notification is provided if the campaign is terminated as a result of the evaluation].
Next, in view of compact prosecution only further analysis per the Berkheimer Memo dated April 19, 2018 is being conducted as the following additional elements would be readily apparent as generic to a person having ordinary skill in the art (hereinafter PHOSITA), in other words analysis is similar to Berkheimer claim 1 and not claims 4-7 where there was "a genuine issue of material fact in light of the specification," nevertheless the Examiner provides citation to one or more publications as noting the well-understood, routine, conventional nature of machine learning as follows:
(i) Chandramouli, Patent: US 8,442,683 note para. [0005]-[0007] and [0029]-[0033]; (ii) Lee, Pub. No.: US 2002/0107926 note para. [0020]; (iii) Kwok, Pub. No.: US 2002/0150295 note para. [0015]; (iv) Teller, Pub. No.: US 2004/0133081 [0236]-[0238]; (v) Agrawal and Srikant Patent No.: US 6546389 note "As recognized herein, the primary task of data mining is the development of models about aggregated data. Accordingly, the present invention understands that it is possible to develop accurate models without access to precise information in individual data records."; (vi) Deshpande et al., Pub. No.: US 2015/0134413 [0046] Using the target and input features, in step F3 of FIG. 1, a plurality of forecasting models are built for a product or a product category, a location, and a time window. A plurality of forecasting models can be built using existing machine learning based methods and/or time-series forecasting methods, and using the standard training-testing-validation methods. In an exemplary embodiment, only the highest quality models with high quality (high accuracy, precision, recall, etc.) are retained.; [0078] The processing system forecasting engine 202 can also include a forecasting model building engine 224 and a forecast calculation engine 226. In the model building stage, target and input features based on a customer or a customer segment's past data are used to train, test, and validate different types of forecasting models using machine learning and/or time series forecasting based approaches. Individual models are retained depending on the performance. The output of plurality of these retained models can then be fused into a single model 228. The fusion can be based on a rule-based approach or by assigning weights to individual model and combining those using ranking or combination techniques." (vii) Wei et al., Pub. No.: US 2015/0235260 [0080] Then, analysis module 532 may determine one or more predefined model(s) 546 based on event data 538 and the one or more targeting criteria. For example, analysis module 532 may use training and testing subsets of this information to generate one or more machine-learning models. The one or more predefined model(s) 546 may allow estimates of the number of future events to be determined for terms 544 in the one or more targeting criteria 542.; (viii) Beatty, Pub. No.: US 2012/0166267 see [0177] note "the prediction of conversion rate is performed by a machine-learning system that is trained using historical purchase data available to the ad system. The training set contains instances of purchase/no purchase decisions and many data points about the (user, context, offer). For example, the training examples might contain the following data points about the offer that was made to a user: price of offer, % discount of offer, popularity of merchant, time of day, gender of user, income of user, interests of user, websites visited by user, categories of websites visited by user, search queries by user, category of business, number of friends that had purchased the offer, "closeness" of friends that had purchased the offer, physical distance between the user's home and the business, physical distance between the user's workplace and the business, the "cluster id" of the user (generated by a clustering algorithm that placed, and users into clusters based on similar attributes of preferences)."
Therefore the claims here fail to contain any additional element(s) or combination of additional elements that can be considered as significantly more and the claims are rejected under 35 U.S.C. 101 for lacking eligible subject matter.
Claim Rejections - 35 USC § 103
4. 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 set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied 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 1-18 are rejected under 35 U.S.C. 103(a) as being unpatentable over Achin et al. (Pub. No.: US 2018/0046926) referred to hereinafter as Achin, in view of Kay H. Brodersen. Fabian Gallusser. Jim Koehler. Nicolas Remy. Steven L. Scott. "Inferring causal impact using Bayesian structural time-series models." Ann. Appl. Stat. 9 (1) 247 - 274, March 2015. https://doi.org/10.1214/14-AOAS788 referred to hereinafter as Brodersen.
As per claim 1, Achin teaches a non-transitory medium with instructions stored thereon that, when executed by a processor of a computing device, cause the computing device to perform operations comprising (see [0477]):
(a) obtaining a dataset that includes a series of values, in temporal order, that are indicative of performance of a company over an interval of time (see Fig. 9 and its associated disclosure; [0119]; [0334]; [0355]-[0356]); segmenting the dataset into (see [0191]; [0192] “the exploration engine 110 suggests data partitions”; [0193]; [0334]) —
(i) a first dataset corresponding to a first period of time […], wherein the first period of time is representative of a subset of the interval of time (see Fig. 9 and its associated disclosure; [0014]; [0119]-[0120]; [0335]; [0345]-[0354]), and
(c) applying the machine learning model to the second dataset, so as to produce a third dataset that is indicative of predicted performance during the second period of time in the absence of the advertising campaign (see Fig. 10 and its associated disclosure; [0014]; [0030]; [0333]-[0334]; [0344]; [0346]);
(e) […] falls beneath a threshold, generating a notification that prompts termination of the advertising campaign (see [0245]-[0247]; [0298]; [0386]; [0404]).
Achin suggests predicting see Fig. 9 and its associated disclosure; [0014]; [0119]-[0120]; [0356]; [0368], however Achin expressly does not teach (a*) […] preceding an introduction of an advertising campaign […] (ii) a second dataset corresponding to a second period of time over which the advertising campaign occurs, wherein the second period of time is representative of another subset of the interval of time; (b) providing the first dataset, but not the second dataset, to a machine learning algorithm that uses the first dataset, to train a machine learning model to predict performance of the company in the absence of the advertising campaign; (d) causing digital presentation of the second and third datasets on an interface as separate traces, so as to visually and programmatically indicate a difference between performance of the company with the advertising campaign and predicted performance of the company without the advertising campaign (see Fig. 1 and its associated disclosure).
Brodersen teaches (a*) […] preceding an introduction of an advertising campaign […] (see Pgs. 248-249; Pg. 261 “To simulate the effect of advertising, the post-intervention portion of the preceding series was multiplied by 1+e, where e (not to be confused with ε) represented the true effect size specifying the (uniform) relative lift during the campaign period. An example is shown in Figure 3(a)”)
(ii) a second dataset corresponding to a second period of time over which the advertising campaign occurs, wherein the second period of time is representative of another subset of the interval of time (see pgs. 248-249; note Fig. 1 and its description on pg. 249);
(b) providing the first dataset, but not the second dataset, to a machine learning algorithm that uses the first dataset, to train a machine learning model to predict performance of the company in the absence of the advertising campaign (see Figs. 1 and 5 and their associated disclosure; pages 248-249; page 258)
(d) causing digital presentation of the second and third datasets on an interface as separate traces, so as to visually and programmatically indicate a difference between performance of the company with the advertising campaign and predicted performance of the company without the advertising campaign (see pgs. 248-249; note Fig. 1 and its description on pg. 249);
(e) in response to a determination that the difference in performance of the company with the advertising campaign and predicted performance of the company without the advertising campaign […] (see Figs. 1 and 5 and their associated disclosure; pages 248-249; page 258).
Therefore it would be obvious to a PHOSITA before the effective filling date of the invention to modify Achin in view of Brodersen with motivation to segment time series data into subsets utilized for training a predictive model, applying the model, and compare it with the actual performance of advertising campaign, see pg. 248 note “Here, we focus on measuring the impact of a discrete marketing event, such as the release of a new product, the introduction of a new feature, or the beginning or end of an advertising campaign, with the aim of measuring the event’s impact on a response metric of interest (e.g., sales). The causal impact of a treatment is the difference between the observed value of the response and the (unobserved) value that would have been obtained under the alternative treatment, that is, the effect of treatment on the treated”.
As per claim 2, Achin in view of Brodersen teaches the claim limitations of claim 1. Achin teaches wherein the interface also includes the first dataset that is presented as a trace (see [0187]).
As per claim 3, Achin in view of Brodersen teaches the claim limitations of claim 1. Achin teaches wherein the operations further comprise: tuning the machine learning model for the company in an autonomous manner using a statistical modeling technique (see [0119]; [0135]; [0255]; [0264]).
As per claim 4, Achin in view of Brodersen teaches the claim limitations of claim 3. Achin suggests see [0174], however Achin expressly does not teach wherein the statistical modeling technique is a Bayesian structural time series. Brodersen teaches wherein the statistical modeling technique is a Bayesian structural time series (see pg. 251 “we use a fully Bayesian approach to inferring the temporal evolution of counterfactual activity and incremental impact. One advantage of this is the flexibility with which posterior inferences can be summarised.”; pg. 252).
Therefore it would be obvious to a PHOSITA before the effective filling date of the invention to modify Achin in view of Brodersen with motivation to use structural time-series models as they are flexible and modular, pg. 252.
As per claim 5, Achin in view of Brodersen teaches the claim limitations of claim 1. Achin teaches wherein the machine learning model includes one or more state variables that, as part of an inferencing operation, are summed in a weighted manner to establish predicted performance (see [0146]).
As per claim 6, Achin in view of Brodersen teaches the claim limitations of claim 5. Achin teaches wherein the machine learning model includes separate state variables for trend, seasonality, and regression, and wherein for each state variable, a corresponding weight is learned through analysis of the first dataset provided to the machine learning algorithm for training purposes (see [0049]; [0246]; [0346]; [0417]-[0419]).
As per claim 7, Achin in view of Brodersen teaches the claim limitations of claim 1. Achin suggests [0469], however Achin expressly does not teach wherein the operations further comprise: employing a Monte Carlo algorithm to find a posterior distribution of an output produced by the machine learning model upon being applied to the second dataset, wherein the Monte Carlo algorithm produces, as output, a sequence of random samples; and using the sequence of random samples to estimate integrals with respect to a target distribution, thereby computing expected values for a key performance indicator by which performance is measured.
Brodersen teaches wherein the operations further comprise: employing a Monte Carlo algorithm to find a posterior distribution of an output produced by the machine learning model upon being applied to the second dataset, wherein the Monte Carlo algorithm produces, as output, a sequence of random samples; and using the sequence of random samples to estimate integrals with respect to a target distribution, thereby computing expected values for a key performance indicator by which performance is measured (see pg. 251 note “The approach described in this paper inherits three main characteristics from the state-space paradigm […] Third, we use a regression component that precludes a rigid commitment to a particular set of controls by integrating out our posterior uncertainty about the influence of each predictor as well as our uncertainty about which predictors to include in the first place, which avoids overfitting. The remainder of this paper is organised as follows. Section 2 describes the proposed model, its design variations, the choice of diffuse empirical priors on hyperparameters, and a stochastic algorithm for posterior inference based on Markov chain Monte Carlo (MCMC).”; pgs. 265-269).
Therefore it would be obvious to a PHOSITA before the effective filling date of the invention to modify Achin in view of Brodersen with motivation to reuse the samples from the posterior to obtain credible intervals for all summary statistics of interest. Such statistics include, for example, the average absolute and relative effect caused by the intervention as well as its cumulative effect., pg. 269.
As per claim 8, Achin in view of Brodersen teaches the claim limitations of claim 7. Achin teaches wherein the key performance indicator is sales, revenue, virality, relevance, or traffic (see [0342]).
As per claim 9, Achin teaches a method performed by a computer program executing on a computing device, the method comprising (see [0016]):
(a) providing a first dataset that includes a first series of values, that are arranged in temporal order and that are indicative of performance of a company over a first interval of time that precedes an advertising campaign, to a machine learning algorithm that uses the first dataset to train a machine learning model to predict performance of the company in the absence of the advertising campaign (see the rejection above for claim 1 limitation (b));
(b) applying the machine learning model to a second dataset that includes a second series of values, that are arranged in temporal order and that are indicative of performance of the company over a second interval of time over which the advertising campaign occurs, so as to produce an output (see Fig. 10 and its associated disclosure; [0014]; [0030]; [0344]; [0346]);
Achin expressly does not teach (c) applying a Monte Carlo algorithm to the output produced by the machine learning model to obtain a series of random samples distributed across a target probability distribution; and (d) estimating, based on the series of random samples, integrals with respect to the target probability distribution, thereby computing expected values for a key performance indicator by which performance of the company is measured.
Brodersen teaches (c) applying a Monte Carlo algorithm to the output produced by the machine learning model to obtain a series of random samples distributed across a target probability distribution (see pg. 251 note “The approach described in this paper inherits three main characteristics from the state-space paradigm […] Third, we use a regression component that precludes a rigid commitment to a particular set of controls by integrating out our posterior uncertainty about the influence of each predictor as well as our uncertainty about which predictors to include in the first place, which avoids overfitting. The remainder of this paper is organised as follows. Section 2 describes the proposed model, its design variations, the choice of diffuse empirical priors on hyperparameters, and a stochastic algorithm for posterior inference based on Markov chain Monte Carlo (MCMC).”; pgs. 265-269); and
Brodersen teaches (d) estimating, based on the series of random samples, integrals with respect to the target probability distribution, thereby computing expected values for a key performance indicator by which performance of the company is measured (see pg. 251 note “The approach described in this paper inherits three main characteristics from the state-space paradigm […] Third, we use a regression component that precludes a rigid commitment to a particular set of controls by integrating out our posterior uncertainty about the influence of each predictor as well as our uncertainty about which predictors to include in the first place, which avoids overfitting. The remainder of this paper is organised as follows. Section 2 describes the proposed model, its design variations, the choice of diffuse empirical priors on hyperparameters, and a stochastic algorithm for posterior inference based on Markov chain Monte Carlo (MCMC).”; pgs. 265-269).
Therefore (as it applies to claim limitations (c) and (d)) it would be obvious to a PHOSITA before the effective filling date of the invention to modify Achin in view of Brodersen with motivation to reuse the samples from the posterior to obtain credible intervals for all summary statistics of interest. Such statistics include, for example, the average absolute and relative effect caused by the intervention as well as its cumulative effect., pg. 269.
As per claim 10, Achin in view of Brodersen teaches the claim limitations of claim 9. Achin teaches wherein the target probability distribution corresponds to the second interval of time (see pg. 248 note “compute the posterior distribution of the counterfactual time series given the value of the target series”; pg. 249; pg. 258).
Therefore it would be obvious to a PHOSITA before the effective filling date of the invention to modify Achin in view of Brodersen with motivation to reuse the samples from the posterior to obtain credible intervals for all summary statistics of interest. Such statistics include, for example, the average absolute and relative effect caused by the intervention as well as its cumulative effect for the second interval of time, pg. 269.
As per claim 11, Achin in view of Brodersen teaches the claim limitations of claim 9. Achin teaches wherein the Monte Carlo algorithm is based on a Markov chain Monte Carlo approach to sampling from the target probability distribution (see pg. 251 note “The approach described in this paper inherits three main characteristics from the state-space paradigm […] Third, we use a regression component that precludes a rigid commitment to a particular set of controls by integrating out our posterior uncertainty about the influence of each predictor as well as our uncertainty about which predictors to include in the first place, which avoids overfitting. The remainder of this paper is organised as follows. Section 2 describes the proposed model, its design variations, the choice of diffuse empirical priors on hyperparameters, and a stochastic algorithm for posterior inference based on Markov chain Monte Carlo (MCMC).”; pgs. 265-269).
Therefore it would be obvious to a PHOSITA before the effective filling date of the invention to modify Achin in view of Brodersen with motivation to reuse the samples from the posterior to obtain credible intervals for all summary statistics of interest. Such statistics include, for example, the average absolute and relative effect caused by the intervention as well as its cumulative effect., pg. 269.
As per claim 12, Achin in view of Brodersen teaches the claim limitations of claim 9. Achin teaches wherein the machine learning model includes one or more state variables that, as part of an inferencing operation, are summed in a weighted manner to establish predicted performance (see [0146]).
As per claim 13, Achin in view of Brodersen teaches the claim limitations of claim 12. Achin teaches wherein for each state variable, a corresponding weight is learned through analysis of the first dataset provided to the machine learning algorithm as part of a training operation (see [0231]), Achin expressly does not teach […] in which a spike-and-slab prior is used for each state variable to allow the machine learning model to regularize and perform feature selection. Brodersen teaches […] in which a spike-and-slab prior is used for each state variable to allow the machine learning model to regularize and perform feature selection (see pgs. 248-249 note “framework of our model allows us to choose from among a large set of potential controls by placing a spike-and-slab prior on the set of regression coefficients and by allowing the model to average over the set of controls [George and McCulloch (1997)]. We then compute the posterior distribution of the counterfactual time series given the value of the target series in the pre-intervention period, along with the values of the controls in the postintervention period”).
Therefore it would be obvious to a PHOSITA before the effective filling date of the invention to modify Achin in view of Brodersen with motivation to use spike and slab methodology for performing feature selection, pg. 248.
As per claim 14, Achin in view of Brodersen teaches the claim limitations of claim 13. Achin teaches wherein for each state variable, a corresponding spike-and-slab prior is representative of a generative model in which that state variable either attains a fixed value or is drawn toward another value (see pg. 248 note “framework of our model allows us to choose from among a large set of potential controls by placing a spike-and-slab prior on the set of regression coefficients and by allowing the model to average over the set of controls [George and McCulloch (1997)]. We then compute the posterior distribution of the counterfactual time series given the value of the target series”; pg. 256 “When faced with many potential controls, we prefer letting the model choose an appropriate set. This can be achieved by placing a spike-and-slab prior over coefficients [George and McCulloch (1993, 1997), Polson and Scott (2011), Scott and Varian (2014)]. A spike-and-slab prior combines point mass at zero (the “spike”), for an unknown subset of zero coefficients, with a weakly informative distribution on the complementary set of nonzero coefficients (the “slab”). Contrary to what its name might suggest, the “slab” is usually not completely flat”; pg. 257).
As per claim 15, Achin in view of Brodersen teaches the claim limitations of claim 9. Achin teaches further comprising: receiving input that is indicative of a selection, made by a user through an interface, of the first and second datasets or another dataset of which the first and second datasets are a part; and obtaining, in response to said receiving, the first and second datasets or the other dataset (see [0179]).
As per claim 16, Achin in view of Brodersen teaches the claim limitations of claim 15. Achin teaches wherein the first and second datasets or the other dataset are acquired via a Representational State Transfer (REST) application programming interface (API) or a database connector (see [0179]; [0284]; [0292]).
As per claim 17, Achin in view of Brodersen teaches the claim limitations of claim 9. Achin teaches further comprising: causing digital presentation of an interface through which a user is able to directly upload one or more files that include the first and second datasets (see [0223]).
As per claim 18, Achin in view of Brodersen teaches the claim limitations of claim 17. Achin teaches wherein the one or more files are comma-separated value (CSV) files or spreadsheet files (see [0481]).
Response to Applicant’s Arguments
4. Regarding Interview, note the Examiner Interview Summary of record 06/20/2025.
Regarding 101, the “Aplicant submits that the claims are not directed to an abstract idea and, even if they were, contain significantly more than any alleged judicial exception. Accordingly, Applicant posits that the claims should be deemed eligible under 35 USC 101 when properly evaluated under the 2019 Revised Patent Subject Matter Eligibility Guidance ("2019 PEG").” However, the Examiner respectfully disagrees.
The Applicant alleges that under step 2A prong one invoking “certain methods of organizing” in view of BRI in light of the as-filed specification is an oversimplification because “Independent claim 1 does not merely organize human activity. It provides a specific, technical solution to a technical problem, namely, how to isolate the effect of an advertising campaign on company performance using a structured, machine learning based modeling framework”. However, the determination is based on BRI in light of the as-filed specification based on abstract recitation not additional elements, as such, the argument in view of machine learning and notification is misplaced. Further there is a clear distinction between using additional elements as high level tools and improving a technology. Thus, the Applicant’s arguments under prong one are unpersuasive.
Next, under prong two, “Even if independent claim 1 were deemed to recite an abstract idea, Applicant submits that its additional elements integrate the abstract idea into a practical application.” Although the Examiner has not invoked mental processes grouping under prong one the Applicant argues that segmenting a dataset temporally is data transformation which is part of machine learning under prong two. However, every machine learning model is trained on supplied and segmentation and/or transformation of said data is based on what is the goal or objective which is what the argued notification outputs based on poor performance of marketing campaign which is measured based on extrapolation of data in absence of marketing campaign compared with plotting of data with marketing campaign and if the difference between the two exceeds some arbitrarily set KPI such as threshold lift then marketing campaign is concluded or terminated, for instance note that the court has already ruled collecting, displaying, and manipulating data (Int. Ventures v. Cap One Financial). This is an abstract idea because (i) the process is similar to collecting information, analyzing it, and displaying certain results of the collection and analysis (Electric Power Group) and (ii) Collecting and analyzing information to detect misuse and notifying a user when misuse is detected (FairWarning). Thus, the Applicant’s arguments are unpersuasive and rather than being akin to McRO (which has no correlation with instant application as the inventions and facts are different), the claims are more analogous to the ones noted above.
Lastly regarding step 2B, the Applicant is reminded that the evaluation is that of additional element(s) singularly and in-combination, not abstract recitation. Thus, as already explained the additional elements are recited at a high level of generality and as already noted above manipulation of data per Int. Ventures v. Cap One Financial and notification of a result or outcome or a decision per notifying a user when misuse is detected is insufficient to supply significantly more.
Regarding 103, the Examiner respectfully disagrees with very limited characterization of the cited prior art references. The Applicant generally discusses each reference individually and then appears to attack each reference individually. Indeed a PHOSITA would be able to properly construe the rejection as relied upon especially Brodersen as depicting the performance of the marketing campaign and comparing it with simultaneously plotted a do-nothing scenario, i.e. in absence of marketing campaign. Indeed the primary reference teaches when certain KPIs are not met the decision making party can be informed/notified/alerted to terminate and/or make adjustments. The rejection has been updated in view of filed claim amendments, as such, the Applicant is requested to note the same, especially note Broderson Figs. 5-7 and their associated disclosure. Therefore, the Examiner respectfully disagrees and maintains the rejection.
Conclusion
5. The prior art made of record and not relied upon is considered pertinent to applicant's disclosure and all the references on PTO-892 Notice of Reference Cited should be duly noted by the Applicant as they can be subsequently used during prosecution, at least note the following:
*Being noted initially
- WO2005/106656A2 note “The invocation of the link to the cumulative lift chart causes display of a cumulative lift chart. The invocation of the link to the cumulative lift chart causes display of a non-cumulative lift chart. A user is enabled to choose interactively at least one performance criterion change or transformation or interaction of variables to improve model validation process. The user interface enables the user to select at least one machine automated model development process applied to the entire dataset for a validated model process. The user interface enables the user to point and click to cause display of information about the performance of the validated model process applied to the entire set of historical data. The information about the model performance for two independent data subsets includes at least one of: a statistical report card with a link to the statistical report chart, a cumulative lift chart with a link to the cumulative lift chart, a non-cumulative lift chart with a link to the non-cumulative lift chart. The invocation of the link to the statistical report card causes display of the statistics of model process validation. The invocation of the link to the cumulative lift chart causes display of a cumulative lift chart. The invocation of the link to the cumulative lift chart causes display of a non- cumulative lift chart. The final model and the model process validation results are stored persistently. In general, in another aspect, a machine-based method includes, in connection with a project in which a user generates a predictive model based on historical data about a system being modeled, for example, displaying to a user a lift chart, monotonicity, and concordance scores associated with each step in a step-wise model fitting process. Implementations may include one or more of the following features. The user is enabled to observe changes in the fit of the model as variables associated with the data are added or removed from a predictor set of the variables. The user is enabled to terminate the fitting of the model when the fitting process reaches an optimal point. In general, in another aspect, a machine-based method includes, for a predictive model based on historical data about a system being modeled, generating measures of the fit of the model to samples of the data, the fit measures being generated separately in percentile segments.”
*Previously noted
- CN109886747 see Abstract “Embodiments of the present invention provide a kind of Method for Sales Forecast method, medium, Method for Sales Forecast device and calculate equipment. The Method for Sales Forecast method includes: the history sales volume time series obtained in first time section; By history sales volume time series be input on ordinary days Method for Sales Forecast model to obtain the prediction sales volume time series in the second time interval; By advertising campaign information input to promotion Method for Sales Forecast model to obtain the promotion sales volume data of promotion period node; Wherein, promotion period node is located in the second time interval; It will predict that the prediction sales volume data for corresponding to promotion period node in sales volume time series replace with promotion sales volume data. Method of the invention can be realized the integrated Method for Sales Forecast to steady sales volume on ordinary days and promotion peak value sales volume, have stronger generalization ability and adaptive adjustment capability by establishing Method for Sales Forecast model on ordinary days and promotion Method for Sales Forecast model.”
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DIPEN M PATEL whose telephone number is (571)272-6519. The examiner can normally be reached Monday-Friday, 08:30-17:00 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, Waseem Ashraf can be reached on (571)270-3948. 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: