DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This Office Action is responsive to Applicant's amendment filed on 26 June 2026. Applicant’s amendment on 26 June 2026 amended Claims 1, 11, and 17-20. Currently Claims 1-7, 11-13 and 15-24 are pending and have been examined. Claims 21-24 are newly presented. Claims 9 and 10 have been canceled. Claim 8 and 14 were previously canceled. The Examiner notes that the 101 has been maintained. Claims 11-13 and 18 have been objected to with respect to the prior art rejection.
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 26 June 2026 has been entered.
Claim Objections
The Examiner notes that the prior art rejections for claims 11, 12, 13, and 18 have been withdrawn because it is not viewed that it would be obvious when combining the number of previously cited references. It is viewed that the claims would be allowable once the remaining rejections are overcome as they still retain their 101 rejection.
Response to Arguments
Applicants’ arguments filed 26 June 2026 have been fully considered but they are not persuasive.
The Applicant argues on pages 11-12 that claim 1 is directed to providing computing resources for a website based on a segment return value an operation that cannot be performed in the human mind. Applicant argues per MPEP 2106.04 (a) and SRI Int’l v. Cisco Systems that claims do not recite a mental process when the human mind is not equipped to perform the claim limitations, and contends that processing time series data for a plurality of users operates at a scales far beyond what any human could practically perform mentally or when pen and paper. Applicant further argues that the Examiner improperly focused only on the “identifying” and “computing” steps while ignoring the claim as a whole.
The Examiner respectfully disagrees.
Applicant's argument that the claimed invention is not a mental process because it operates at a scale beyond human capability is not persuasive because it conflates the scale of data processing with the nature of the underlying conceptual steps being performed. The broadest reasonable interpretation of claim 1 encompasses a process of: (1) identifying a segment of users based on time series data; (2) computing first and second subsets based on a threshold number of interactions; (3) generating a segment return value by combining return values for each subset; and (4) providing computing resources based on that value. The core conceptual steps of identifying user groups based on interaction frequency, making a prediction about future behavior from historical data, and then deciding to allocate resources accordingly are precisely the type of observations, evaluations, and judgments that courts have recognized as performable mentally or with pen and paper. While Applicant invokes SRI Int'l v. Cisco Systems and the principle that claims operating at a scale where "the human mind is not equipped" to perform the steps are not mental processes, that principle is inapplicable here because the claims do not recite the type of inherently machine-dependent operations such as monitoring network packets in real time, performing GPS pseudorange calculations, or executing data encryption protocols that SRI Int'l, SiRF Tech., and the related cases addressed. Rather, the claims at their broadest recite classifying users into groups, calculating ratios of return rates from historical data, and provisioning resources all of which are conceptually analogous to the steps in Electric Power Group v. Alstom, where data analysis steps "recited at a high level of generality" were found to be performable mentally, and in In re Killian, where determining eligibility status by collecting information and making determinations from multiple sources was likewise a mental process.
Furthermore, Applicant's argument that the final step of "providing computing resources" removes the claims from the mental process grouping is not persuasive because, as the USPTO guidance and controlling case law confirm, a claim can recite a mental process even when claimed as being performed on a computer, and adding a final step of acting on the result of a mental process does not take the underlying abstract idea out of the mental processes grouping. See MPEP 2106.04(a)(2)(III)(C); Mortgage Grader, Inc. v. First Choice Loan Servs. The claim limitations singled out by Applicant the identifying step and the computing step are, under their broadest reasonable interpretation, the heart of the claimed method, and they remain performable as a conceptual matter without the recitation of any machine that is integral to those steps in the way required by the cited non-mental-process cases. Accordingly, the claims continue to recite a mental process abstract idea under Step 2A, Prong One, and the rejection is therefore maintained.
The Applicant argues on pages 12-13 that “ the claim 1 recites limitations that cannot practically be performed in the human mind or with pen and paper and is an operation performed on time series data that exists only because it is generated and collected by computer component interacting with a web browser. The segment is identified from time series data that is generated by the interaction of a plurality of users with a website through a web browser-for example, a visit to a URL, leaving a URL, a referrer URL, a click on a hyperlink, or adding or removing a product from a virtual shopping cart, each recorded with the date and time of the interaction (see Specification at paragraph [0093]). A human mind is not equipped to observe, capture, and aggregate the browser-level interaction records of a plurality of users across a first time period, nor to partition those users into a segment on the basis of that machine-collected time series data.
The Examiner respectfully disagrees.
With respect to the argument the Examiner notes that the Applicant's argument directed specifically to the "identifying" limitation. Applicant contends that the identifying step cannot be performed mentally because it depends on time series data that is generated and collected by computer components interacting with a web browser for example, URL visits, hyperlink clicks, and shopping cart interactions, each recorded with a date and timestamp. This argument is unpersuasive because it conflates the source or origin of the underlying data with the nature of the conceptual operation being claimed. The "identifying" limitation as claimed requires identifying a segment of a plurality of users for a first time period based on time series data. Under its broadest reasonable interpretation, this step recites nothing more than examining a collection of user interaction records and grouping users into a segment based on shared characteristics precisely the type of observation, evaluation, and judgment that courts have consistently identified as a mental process. A human analyst who is provided with the same interaction records could perform this segmentation using pen and paper: reviewing each user's interactions with a website during a given time period and placing those who share the relevant attribute into a common group. The fact that the underlying data is generated by browser-level events does not alter the nature of the identifying step itself, which is recited at a high level of generality with no structural limitation on how the segmentation is accomplished or what machine is integral to performing it. See Electric Power Group v. Alstom, (data analysis steps recited at a high level of generality can practically be performed in the human mind regardless of the volume or digital nature of the data involved); In re Killian, (collecting information from multiple sources and making determinations based on that information is a mental process even when multiple data sources are involved).
Moreover, Applicant's reliance on the specification's description of the types of interactions captured in the time series data does not rescue the claim limitation, because the claim itself does not require any particular machine-dependent mechanism for collecting or processing those interactions. See Symantec, 838 F.3d at 1318 (although claimed as computer-implemented, steps that can be performed by a human mentally or with pen and paper remain a mental process). The "segmentation component" recited in the claim is a generic computer component that is itself recited at a high level of generality and merely serves as a tool for performing the underlying conceptual operation of classifying users it does not structurally or functionally limit the identifying step in a manner that removes it from the mental processes grouping. Accordingly, the identifying limitation continues to recite a mental process abstract idea, and this argument does not overcome the rejection.
The Applicant argues on pages 13-14 that the computing limitation which requires partitioning users into a first subset of those exceeding a threshold interaction count and a second subset of those falling below it cannot practically be performed in the human mind or with pen and paper. Applicant contends that because computing the two subsets requires tallying each individual user’s interaction count across a range of machine-collected, browser-level interaction records and then comparing the count against a threshold to assign each user to the appropriate subset, the operation is dependent on data that exists only because it is generated and collected by computer components interacting with a web browser. Applicant therefore concludes that the human mind is not equipped to observe, capture, aggregate, and partition users on the basis of such data, and that this step is accordingly not a mental process.
The Examiner respectfully disagrees.
With respect to the argument the Examiner notes that the Applicant's argument is unpersuasive because it conflates the origin of the underlying data with the nature of the conceptual operation being claimed. Under its broadest reasonable interpretation, the computing limitation requires nothing more than examining user interaction records, counting each user's interactions against a threshold number, and partitioning users into one of two groups depending on whether they exceed or fall below that threshold a binary classification based on a numerical comparison that is precisely the type of observation, evaluation, and judgment that a human can perform with pen and paper when provided the relevant records. See Electric Power Group v. Alstom,; CyberSource Corp. v. Retail Decisions, Inc. The correct legal standard under MPEP 2106.04(a)(2)(III) is whether the claimed limitation itself not the data upon which it operates can practically be performed in the human mind or with a physical aid, and the computing limitation as recited at a high level of generality satisfies that standard regardless of whether the underlying data is digitally generated. The generic recitation of a "prediction component" does not alter this analysis as it is itself recited at a high level of generality and serves only as a nominal instruction to apply the abstract classification using a generic computer tool. The rejection are therefore maintained.
The Applicant argues on page 14 that the generating limitation which requires computing separate numerical return values for the first and second subsets and combining them into a segment return value cannot practically be performed in the human mind or with pen and paper. Applicant argues “forming a judgement about predicted future interactions,” does not recite a subjective estimate, opinion, or judgement but rather the computation of precise numerical representations of predicted future returning users derived from machine-collected, browser-level time series data.
The Examiner respectfully disagrees.
This argument is unpersuasive for the same fundamental reason addressed in the argument above the Applicant again conflates the volume and digital origin of the underlying data with the nature of the conceptual operation being performed by the claim limitation itself. Under its broadest reasonable interpretation, the computing limitation requires examining user interaction records, counting each user's interactions against a threshold number, and partitioning users into one of two groups depending on whether they exceed or fall below that threshold. This is conceptually equivalent to a sorting or classification operation that is precisely the type of comparison, evaluation, and judgment that a human can perform with pen and paper when given a set of records. The fact that the records happen to be browser-level interaction data collected by a computer does not transform the nature of the claimed step. See Electric Power Group v. Alstom (collecting, analyzing, and displaying information steps recited at a high level of generality remain mentally performable regardless of the scale or digital nature of the underlying data); CyberSource Corp. v. Retail Decisions, Inc., (a step of constructing a map of credit card transactions was a mental process performable with pen and paper even though the underlying data was electronically generated, because a human could write down the relevant transactions and perform the comparison). Indeed, the specification itself describes the first subset as including users who interact with the content channel more than a threshold number of times, and the second subset as the complement a straightforward binary partition based on a count comparison that is conceptually no different from a human reviewing a list of customer visits and placing frequent visitors in one column and infrequent visitors in another.
Furthermore, Applicant's argument that this step "could not practically be performed as a mental observation, evaluation, or judgment" because the data "exists only because it is generated and collected by computer components" is not the correct legal standard. The proper inquiry under MPEP 2106.04(a)(2)(III) and controlling case law is whether the claimed limitation itself not the data upon which it operates can practically be performed in the human mind or with a physical aid such as pen and paper. See Synopsys, Inc. v. Mentor Graphics Corp; In re Killian. The computing limitation recites only that users whose interaction count exceeds a threshold are placed in the first subset, and the remainder constitute the second subset a binary classification based on a numerical comparison that a human can readily perform with pen and paper if provided the relevant records. The generic recitation of a "prediction component" does not alter this analysis because, as the case law confirms, a claim can recite a mental process even when nominally assigned to a computer component, where, as here, that component is itself recited at a high level of generality and serves only as an instruction to apply the abstract idea using a generic computer tool. See Mortgage Grader, Inc. v. First Choice Loan Servs; Intellectual Ventures I LLC v. Symantec Corp. Accordingly, the computing limitation continues to recite a mental process abstract idea, and this argument does not overcome the rejection and the rejection is maintained.
The Applicant argues on page 15 that Applicant argues that the "providing computing resources" limitation is not a mental act in any respect, as allocating bandwidth, server, or cloud-based service resources to a website for a future time period is a concrete technological action affecting computing infrastructure that cannot be performed in the human mind or with pen and paper.
The Examiner respectfully disagrees.
With respect to the argument the Examiner notes that regarding the "providing computing resources" limitation, the Examiner acknowledges that allocating bandwidth, server, or cloud-based service resources to a website is not itself a mental act. However, the "providing computing resources" limitation, while not itself a mental act, does not rescue the claim from the abstract idea finding because it functions as post-solution activity that is nominally appended to the abstract steps. See Parker v. Flook, ("Post-solution activity, no matter how conventional or obvious in itself, can not transform an unpatentable principle into a patentable process"); Ultramercial, Inc. v. Hulu, LLC. The "providing computing resources" step is recited entirely at a result-oriented, functional level it simply directs that resources be allocated based on the output of the abstract computation and does not recite any particular mechanism, technical implementation, or structural limitation that would meaningfully integrate the abstract idea into a practical application at Step 2A, Prong One. See Internet Patents Corp. v. Active Network, Inc., (a limitation describing "the effect or result dissociated from any method by which" the result is accomplished does not provide a meaningful limitation).
Applicant also challenges the Examiner’s characterization of the claims as directed to “estimating segment size during time periods of content to provide customized content,” arguing that the description is untethered from the actual claim language because the claims do not recite providing customized content at all. Examiner acknowledges Applicant's point that the amended claims, as currently presented, recite providing computing resources rather than providing customized content. The Examiner notes this characterization and clarifies for the record that the abstract idea is more precisely identified as: collecting and analyzing time series data representing user interactions with a website; identifying user segments and subsets based on interaction frequency relative to a threshold; and computing numerical predictions of future user interaction by calculating and combining return values for those subsets all of which collectively constitute abstract mental processes and mathematical concepts as established in the responses to Arguments 1 through 3. This clarification, however, does not alter the substance of the rejection, as the abstract idea identified is fully supported by the claim language regardless of whether the downstream application is characterized as providing customized content or providing computing resources, the 101 rejection is therefore maintained.
The Applicant argues on pages 15-17 that “even assuming claim 1 recites a judicial exception, the claim as a whole integrates that exception into a practical application under Step 2A, Prong Two because the "providing computing resources" limitation requiring allocation of bandwidth, server, or cloud-based service resources based on the segment return value” reflects a concrete improvement to the technological process of provisioning computing resources to a website. Applicant grounds this argument in the specification, citing paragraphs [0015]–[0016], [0022], [0027], and [0099]–[0131], and that the claim reflects that improvement by tying the more accurate segment return value directly to the allocation of specific computing resources”.
The Examiner respectfully disagrees.
The Examiner notes that the rejection of claims 1, 17, and 20 is maintained. While the Examiner acknowledges Applicant's citation to the specification's description of improved resource allocation and has carefully considered the two-step improvements analysis required under MPEP § 2106.04(d)(1) and the precedential Ex Parte Desjardins decision, the argument remains unpersuasive for the following reasons.
The improvements consideration under Step 2A, Prong Two requires a two-part evaluation: first, whether the specification provides sufficient details such that one of ordinary skill in the art would recognize the claimed invention as providing an improvement to the functioning of a computer or to another technology or technical field; and second, whether the claim itself reflects that disclosed improvement. See MPEP § 2106.04(d)(1); Ex Parte Desjardins, Appeal No. 2024-000567 (PTAB Sept. 26, 2025) (precedential). Critically, an improvement to the abstract idea itself such as a more accurate prediction method is not an improvement to technology. See In re Board of Trustees of Leland Stanford Junior University, (an "improved mathematical process" is not an "improved technological process"); 101 Examination Guidance at Section III.A.2.a ("An improvement in the judicial exception itself is not an improvement in the technology").
Turning first to the specification, while the specification does describe the claimed invention as providing more accurate predictions of future user segment sizes than conventional systems, and references the use of those predictions to determine "an amount of bandwidth, servers, cloud-based services, and the like that should be provided for the content channel" (par. [0027]), the improvement described in the specification is fundamentally an improvement to the accuracy of a predictive mathematical calculation not to how computing infrastructure itself operates. The specification does not describe, for example, any new or unconventional mechanism by which bandwidth is allocated, any specific technical protocol by which server or cloud resources are provisioned, or any improvement to the architecture or functioning of the underlying computing systems. The cited paragraphs [0015]-[0016], [0022], and [0027] describe the benefit of a more accurate segment return value and note that this value can inform resource allocation decisions, but the specification does not disclose a technological solution to a technological problem in the computing infrastructure itself. This is analogous to the situation in Trading Technologies Int'l v. IBG, Inc., 921 F.3d 1084, 1093 (Fed. Cir. 2019), where using a computer to provide a trader with more information to facilitate market trades improved the business process of trading but not the computer itself, and is distinguishable from cases such as Enfish, McRO, and Finjan where the specification described how the claimed invention specifically improved how the underlying technology the database, the animation software, or the virus scanner operated at a technical level.
Turning to the claim itself, even assuming the specification could be read to describe a technological improvement, the claim does not reflect any such improvement in its limitations. The "providing computing resources" limitation recites only a result that bandwidth, server, or cloud-based service resources are "allocated for use by the website during the future time period when the segment return value indicates the segment of users will have a repeat interaction with the website" without reciting any particular technical mechanism, architecture, protocol, or process by which those resources are provisioned or allocated. This result-oriented, outcome-based recitation covers any possible way of allocating computing resources and does not confine the claim to a particular technical solution. See Internet Patents Corp. v. Active Network, Inc., 790 F.3d 1343, 1348 (Fed. Cir. 2015); Electric Power Group v. Alstom, 830 F.3d 1350, 1356 (Fed. Cir. 2016). Accordingly, even if the specification described an improvement to resource provisioning technology which it does not at a sufficient level of technical detail the claim itself does not reflect that improvement because it lacks the components or steps that would provide such an improvement. See Intellectual Ventures I LLC v. Symantec Corp., 838 F.3d 1307, 1316 (Fed. Cir. 2016). In sum, what the claims describe is an improvement to the accuracy of a mathematical prediction process, which is an improvement to the abstract idea itself rather than to any computer technology or technical field, and therefore does not integrate the judicial exception into a practical application at Step 2A, Prong Two. The rejection is accordingly maintained.
The Applicant argues on pages 17-19 that “claim 1 not only discloses a technological improvement in the specification but also reflects that improvement in the claim itself, as required by the 2019 PEG Update and Ex Parte Desjardins. Applicant contends that, the present specification explains how computing the segment return value from two subsets achieves more accurate resource provisioning, and claim 1 reflects that improvement by requiring that the more accurate segment return value be used to allocate specific bandwidth, server, or cloud-based service resources to the website for the future time period. Applicant further argues that the Examiner's characterization of the improvement as directed solely to the accuracy of an abstract prediction improperly isolates the prediction step from the resource-provisioning limitation with which it is integrated, and that when the claim is evaluated as a whole and as an ordered combination per MPEP § 2106.05(a) and McRO, the improvement is properly understood as an improvement to the technological process of provisioning computing resources to a website which constitutes an improvement to "another technology or technical field" under MPEP § 2106.05(a) rather than a mere improvement to an abstract computation”.
The Examiner respectfully disagrees.
The Examiner has carefully considered Applicant's argument that the claim as a whole, including the "providing computing resources" limitation evaluated together with the abstract computational steps, reflects a disclosed technological improvement to website resource provisioning. This argument is unpersuasive for the reasons that follow, which distinguish the present claims from the facts of Desjardins in a legally significant manner.
In Desjardins, the specification identified a specific and concrete technological problem of"catastrophic forgetting" in continual learning machine learning systems and described a specific technical mechanism by which the claimed invention solved that problem at the level of how the machine learning model itself operates: by training the model to learn new tasks while actively protecting the values of parameters associated with prior tasks, thereby simultaneously reducing storage capacity requirements and system complexity. Critically, the claims in Desjardins reflected that improvement by reciting the specific technical operation that directly implemented the disclosed technical solution. The improvement in Desjardins was thus an improvement to how the machine learning technology itself functioned in operation, not merely an improvement to the accuracy of a calculation whose output was then applied to a downstream technological action. See Desjardins Memo at 2 ("Such improvements were tantamount to how the machine learning model itself would function in operation and therefore not subsumed in the identified mathematical calculation.").
The present claims are materially distinguishable. Here, the specification does not describe a specific technical problem with how bandwidth allocation, server provisioning, or cloud-based service deployment operates at a technical level, nor does it describe a technical mechanism by which the computing infrastructure itself is improved in its operation. The specification describes a problem with the accuracy of conventional user segment size prediction methods and discloses a mathematical solution computing a segment return value from two subsets rather than from the segment as a whole that produces a more accurate numerical prediction. The specification then notes, at a high level of generality, that this more accurate prediction can be used to determine resource allocation. This is precisely the type of downstream application of an abstract computation to a technological environment that the guidance and case law distinguish from a genuine technological improvement. See Trading Technologies Int'l v. IBG, Inc., (providing more information to facilitate a business process improves the business process, not the computer); Stanford I, (an improved mathematical process yielding better outputs is not an improved technological process). Furthermore, while the Desjardins Memo cautions that examiners "should not evaluate claims at such a high level of generality that potentially meaningful technical limitations are dismissed without adequate explanation," it equally does not require that any result-oriented recitation of a downstream technological action however generic transforms an otherwise abstract claim into an eligible one. The "providing computing resources" limitation, as recited, covers any possible way of allocating any bandwidth, server, or cloud-based service resource to a website and does not specify any particular technical mechanism, protocol, configuration, or architectural arrangement by which that allocation is accomplished. This broad, outcome-based recitation does not reflect a specific technological improvement to computing infrastructure in the way that the claims in Desjardins, Enfish, McRO, BASCOM, and DDR Holdings reflected specific technical solutions to technical problems in those respective technologies. The rejection is accordingly maintained.
The Applicant argues on page 19 that “the Office Action also contends that the Specification describes the alleged improvement only in a conclusory manner (see Office Action, pages 10-11). Applicant respectfully submits the contrary. As discussed above, the Specification does not merely assert an improvement; it explains how the improvement is achieved-by dividing the segment into a first subset and a second subset based on varying levels of historical engagement, computing a separate return value for each subset, and combining the return values to obtain a more accurate segment return value that is then used to provision computing resources (paragraphs [0015]-[0016], [0022], [0027], and [0099]-[0131]). This is precisely the kind of explanation of how the claimed mechanism achieves the improvement that the December Memo and Ex Parte Desjardins identify as sufficient”.
The Examiner respectfully disagrees.
With respect to the argument the Examiner notes that the rejection of claims 1, 17, and 20 is maintained notwithstanding Applicant's argument that the specification explains the improvement in sufficient non-conclusory detail. The Examiner has carefully considered Applicant's citation to paragraphs [0015]-[0016], [0022], [0027], and [0099]-[0131] of the specification and Applicant's reliance on Ex Parte Desjardins and the December 5, 2025 Memorandum. This argument remains unpersuasive because it conflates a detailed explanation of how a more accurate mathematical calculation is achieved with a detailed explanation of how a technology or technical field is improved.
The two-step inquiry under MPEP 2106.04(d)(1) and Desjardins requires, at the first step, that the specification describe the invention such that an improvement to the functioning of a computer or to another technology or technical field would be apparent to one of ordinary skill in the art not merely that the specification explains how a more accurate result is computed. The guidance is explicit that "a bare assertion of an improvement without the detail necessary to be apparent to a person of ordinary skill in the art" is insufficient, and equally that an improvement to the abstract idea itself is not an improvement to technology. See MPEP 2106.04(d)(1); 101 Examination Guidance at Section III.A.2.a; Stanford I.
Examining the cited paragraphs in detail, paragraphs [0015]-[0016] explain that conventional systems "do not accurately predict a size of a future segment of users" and that the claimed system addresses this by "using historical time series data to identify the segment and the subdivisions, making separate predictions for the subdivisions, and making a final segment size estimate based on the separate predictions." Paragraph [0022] states that by "basing the segment return value on the first subset and the second subset, the system provides a more accurate prediction of a size of a future segment of users than conventional techniques can provide." Paragraph [0027] notes that "an example of an efficient allocation of resources is determining an amount of bandwidth, servers, cloud-based services, and the like that should be provided for the content channel based on the segment return value." Paragraphs [0099]-[0131] describe the mathematical formulas and equations by which the segment return value, first return value, and second return value are computed, including ratio calculations, moving average estimators, ARMA models, and ARIMA models.
While the Examiner acknowledges that this is a detailed and technically thorough description, what is described in detail is how the mathematical prediction is computed more accurately not how the computing infrastructure itself is improved in its operation or architecture. Every sentence in the cited paragraphs that relates to an improvement describes an improvement to the accuracy of a numerical prediction. The sole reference to computing resources in the cited paragraphs is paragraph [0027], which states only that the segment return value can be used to determine an amount of bandwidth, servers, or cloud-based services to be provided a single, high-level statement that does not explain any particular technical mechanism, protocol, or architecture by which resource provisioning is improved, does not identify any technical problem in how computing resources are currently allocated, and does not describe how the claimed method changes how the computing infrastructure itself operates. This stands in stark contrast to Desjardins, where the specification provided detailed technical explanation of how the machine learning model itself would function differently in operation specifically, how parameter values associated with prior tasks would be actively protected during subsequent training, how this addressed the well-defined technical problem of catastrophic forgetting at the level of model operation, and how this simultaneously reduced storage requirements and system complexity as concrete, technology-level outcomes. The Desjardins Memo is explicit that what distinguished that specification was that the improvements "were tantamount to how the machine learning model itself would function in operation and therefore not subsumed in the identified mathematical calculation." The present specification's description of resource allocation, by contrast, is entirely downstream of and separate from the mathematical computation and describes only the idea of using a better prediction to make a resource allocation decision which is precisely what the guidance identifies as generally linking a judicial exception to a technological environment rather than integrating it into a practical application. See MPEP 2106.05(h). Accordingly, the specification does not provide the quality of technological improvement disclosure required under Desjardins and MPEP 2106.04(d)(1) with respect to the computing resource provisioning technology, and the rejection is therefore maintained.
The Applicant argues on pages 19-20 that “the Office Action characterizes the "providing computing resources" limitation as a mere instruction to "apply" the abstract idea, as insignificant post-solution activity, and as a field-of-use limitation (see Office Action, pages 12-13; see also pages 16-18). Applicant respectfully traverses. The limitation is not a nominal instruction to apply the result in some unspecified field; it specifies a particular technological action-allocating a bandwidth resource, a server resource, or a cloud-based service resource to the website for the future time period, conditioned on the segment return value indicating a repeat interaction-and is integral to the claimed improvement. This is unlike the bare adjustment of an alarm-limit number in Parker v. Flook, 437 U.S. 584 (1978); see MPEP 2106.05(g). Here, the claimed step improves the provisioning of computing resources to a website and therefore meaningfully integrates any judicial exception into a practical application”.
The Examiner respectfully disagrees.
This argument is unpersuasive for the following reasons.
First, regarding Applicant's contention that the limitation is not a mere "apply it" instruction, the Examiner acknowledges that the limitation recites specific categories of computing resources bandwidth, server, and cloud-based service resources and conditions their allocation on the segment return value indicating a repeat interaction. However, the critical inquiry under MPEP 2106.05(f) and the improvements consideration is not whether the additional element names specific types of resources, but whether the claim as a whole recites a particular solution to a technological problem or merely recites the idea of using an abstract result to achieve a desired outcome. See Internet Patents Corp. v. Active Network, Inc; Electric Power Group v. Alstom. The "providing computing resources" limitation, under its broadest reasonable interpretation, covers any possible way of allocating any bandwidth, server, or cloud-based service resource to a website when a predicted repeat interaction is indicated. It recites no particular technical mechanism, protocol, architecture, or configuration by which the allocation is accomplished. The mere enumeration of categories of computing resources bandwidth, server, cloud-based services does not specify how those resources are provisioned, in what manner the allocation is technically implemented, or what technical constraints govern the allocation process. This is precisely the type of result-oriented recitation that the case law identifies as equivalent to the words "apply it," because it covers any solution to the identified outcome without restriction on how the result is accomplished and without description of the mechanism for accomplishing the result. See MPEP 2106.04(d), subsection III(1); Intellectual Ventures I v. Capital One Financial Corp.,.
Second, regarding Applicant's attempt to distinguish Parker v. Flook, the Examiner notes that while the present limitation does go beyond the bare numerical adjustment at issue in Flook, the relevant legal standard is not whether the limitation exceeds the minimal post-solution activity in Flook, but whether it constitutes a meaningful integration of the judicial exception into a practical application of that exception. The case law establishes a spectrum between the bare alarm-limit adjustment in Flook and the specific, integral technical solutions found in cases such as DDR Holdings, BASCOM, and McRO, and the mere fact that a limitation does more than the Flook adjustment does not establish that it rises to the level of meaningful integration. Here, the "providing computing resources" limitation does not specify how the computing infrastructure is technically affected by or responds to the segment return value, does not recite any particular arrangement of computing components that is integral to achieving the claimed outcome, and does not describe any technical interaction between the prediction mechanism and the resource provisioning system beyond the functional instruction that resources should be allocated when the prediction indicates a repeat interaction. This is analogous to the administration step in Mayo Collaborative Services v. Prometheus Laboratories, 5, which the Supreme Court found to be insignificant post-solution activity even though it specified a concrete medical action, because it merely told doctors to apply the result of the abstract correlation without specifying any particular manner of doing so that would meaningfully limit the claim.
Third, regarding Applicant's contention that the limitation is not a field-of-use limitation, the Examiner acknowledges that the limitation does more than simply identify a field of use by naming categories of computing resources. However, the limitation nonetheless functions to generally link the use of the abstract prediction to the technological environment of website resource management without imposing any meaningful technical constraint on how that linkage is implemented. As the Supreme Court explained in Diamond v. Diehr, a claim directed to a judicial exception cannot be made eligible "simply by having the applicant acquiesce to limiting the reach of the patent for the formula to a particular technological use. The "providing computing resources" limitation, as recited at a high level of functional generality, achieves exactly this: it limits the reach of the abstract prediction computation to the technological environment of website resource provisioning without specifying any particular technical implementation that would distinguish the claimed method from the abstract idea of computing a prediction and using it to make a resource allocation decision. Accordingly, whether characterized as a mere "apply it" instruction, insignificant post-solution activity, or a field-of-use limitation each of which independently supports the rejection the "providing computing resources" limitation does not meaningfully integrate the recited judicial exception into a practical application, and the rejection is therefore maintained.
The Applicant argues on page 20 that “providing computing resources" limitation is not insignificant extra-solution activity because it is not a nominal or tangential addition to the claim but rather the concrete technological action that the entire claimed prediction methodology is designed to enable, imposing a meaningful limit by conditioning the allocation of bandwidth, server, or cloud-based service resources on the segment return value indicating a repeat interaction.
The Examiner respectfully disagrees.
Regarding Applicant's argument that the "providing computing resources" limitation is not insignificant extra-solution activity. The Examiner acknowledges that the limitation does condition the allocation of computing resources on the segment return value indicating a repeat interaction, and that this conditional structure goes beyond a bare, unconditional post-solution step. However, the relevant inquiry under MPEP 2106.05(g) is not simply whether a limitation is conditioned on the output of the abstract idea, but whether the limitation is significant that is, whether it imposes a meaningful limit on the claim such that it is not nominally or tangentially related to the invention and whether it amounts to necessary data output or application that all uses of the recited judicial exception would require. See MPEP 2106.05(g)(2)-(3); Parker v. Flook; OIP Technologies, Inc. v. Amazon.com, Inc.
Here, the "providing computing resources" limitation, while conditioned on the segment return value, is insignificant extra-solution activity for the following reasons. First, the limitation is recited entirely at a functional, outcome-oriented level without any technical detail regarding how the allocation is accomplished. It simply directs that a bandwidth, server, or cloud-based service resource be allocated when a predicted repeat interaction is indicated. This is analogous to the step of "presenting offers to potential customers" in OIP Technologies, which the Federal Circuit found to be insignificant extra-solution activity because it merely gathered and applied data as a necessary input and output to the abstract calculation without imposing any meaningful technical constraint on how the process was carried out. Third, Applicant's characterization of this limitation as "the technological action that the claimed prediction is computed to enable" actually underscores its post-solution character: it is the downstream application of the abstract computation, not an integral technical step that transforms how the computation is performed or how the computing infrastructure operates. See Flook, ("The notion that post-solution activity, no matter how conventional or obvious in itself, can transform an unpatentable principle into a patentable process exalts form over substance."); Ultramercial, Inc. v. Hulu, LLC, the rejection is maintained.
The Applicant argues on pages 20-23 that “claim 1 recites additional elements that, individually and as an ordered combination, amount to significantly more than the alleged judicial exception when considering the claimed combination as a whole and the allocating bandwidth, servers, and cloud-based services to a website is not well-understood, routine, conventional activity.
The Examiner respectfully disagrees.
Regarding Applicant's argument that the ordered combination of generating a segment return value from first and second subsets and providing computing resources based on that value is not well-understood, routine, or conventional, and thus constitutes an inventive concept, this argument is unpersuasive. The Step 2B analysis requires evaluating the additional elements those beyond the judicial exception itself both individually and in combination to determine whether they amount to significantly more than the abstract idea. See MPEP 2106.05; Alice Corp. As established throughout the responses to Arguments 1 through 10, the core of the claimed method identifying user segments, computing subsets based on interaction frequency thresholds, and generating a segment return value by computing and combining return values for those subsets constitutes the recited abstract idea itself. An inventive concept "cannot be furnished by the unpatentable law of nature (or natural phenomenon or abstract idea) itself." Genetic Technologies Ltd. v. Merial LLC; Alice Corp. Accordingly, Applicant cannot establish an inventive concept by pointing to the abstract computational steps themselves. The additional element beyond the abstract idea is the "providing computing resources" limitation, which for the reasons set forth in the responses to Arguments 4, 9, and 10 above amounts to post-solution activity, a field-of-use limitation, and a mere "apply it" instruction that does not meaningfully limit the abstract idea and does not supply an inventive concept even when considered in combination with the abstract computational steps. See Flook; Alice Corp.; OIP Technologies.
Regarding Applicant's challenge to the factual support for the well-understood, routine, and conventional finding, Applicant correctly states that a conclusion that additional elements are well-understood, routine, and conventional must be supported by a factual showing under MPEP 2106.05(d) and Berkheimer v. HP Inc., and that such support must be directed to the claimed combination rather than merely to individual generic components. The Examiner acknowledges Applicant's point that the Office Action's assertion regarding the routine nature of computing resource allocation was not accompanied by a specific case citation. The Examiner hereby supplements the record as follows to provide the required factual support.
First, the specification demonstrates the well-understood, routine, and conventional nature of the additional elements. The specification describes bandwidth, servers, and cloud-based services at paragraph [0027] as known types of resources that can be allocated to a content channel stating that "[a]n example of an efficient allocation of resources is determining an amount of bandwidth, servers, cloud-based services, and the like that should be provided for the content channel based on the segment return value" without providing any technical detail regarding how such allocation is accomplished, thereby indicating that these resource types and their allocation are sufficiently well-known that the specification need not describe the particulars of the allocation mechanism to satisfy 35 U.S.C. 112(a). See MPEP 2106.05(d)(I)(A).
Second, the "providing computing resources" limitation, recited at a high level of generality without any particular technical mechanism for how allocation is accomplished, falls squarely within these recognized categories of well-understood, routine, and conventional computer-implemented activities.
Third, and critically, even if Applicant were correct that the specific combination of allocating computing resources based on a segment return value computed from two subsets were not individually well-understood, routine, and conventional, the combination of additional elements here does not exhibit the type of non-conventional, non-generic arrangement that has been found to supply an inventive concept under BASCOM, DDR Holdings, or Diamond v. Diehr. In those cases, the combination of elements produced a technical improvement in how the underlying technology itself operated filtering content at a specific unconventional location in BASCOM, overriding the conventional sequence of events triggered by a hyperlink click in DDR Holdings, and integrating a mathematical equation into a specific physical industrial process with real-time monitoring in Diehr. Here, the combination of computing a segment return value through a two-subset prediction methodology and then directing that a computing resource be allocated based on that value produces no analogous technical improvement to how the computing infrastructure operates it simply applies an abstract prediction to a conventional resource allocation decision. Accordingly, the additional elements, considered individually and in combination, do not amount to significantly more than the recited abstract idea, and the rejection is maintained.
The remaining Applicant's arguments filed 26 June 2026 have been fully considered but they are moot in view of new grounds of rejection as necessitated by amendment.
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 1–7, 11–13, and 15–24 are rejected under 35 U.S.C. § 101 because the claimed invention is directed to an abstract idea without significantly more. The claims recite an abstract idea falling within the mental processes and mathematical concepts groupings specifically, the collection and analysis of time series data representing user interactions with a website, the identification of user segments and subsets based on interaction frequency relative to a threshold, and the computation of numerical predictions of future user interaction by calculating and combining return values for those subsets. The claims do not include additional elements sufficient to amount to significantly more than the judicial exception because the only substantive additional element beyond the abstract idea the provision of computing resources based on the segment return value is well-understood, routine, and conventional activity in the field, is recited at a high level of generality without any specific technical mechanism for how the allocation is accomplished, and does not supply an inventive concept individually or in combination with the abstract steps.
STEP 1
Claims 1–7, 11-13, and 15–24 are directed to a process/method. Claims 17 and 19 are directed to a non-transitory computer-readable medium (manufacture). Claims 20–24 are directed to a system/machine. Therefore, all claims fall within the statutory categories of invention under 35 U.S.C. 101, and the analysis proceeds to Step 2A.
STEP 2A, Prong One
The independent claims 1, 17, and 20 each recite the following limitations that constitute the recited abstract idea:
First Limitation: "identifying, a segment of a plurality of users for a first time period based on time series data for a website, wherein the time series data includes a series of interactions between the plurality of users and the website via a web browser and wherein the segment includes a portion of the plurality of users interacting with the website during the first time period"
Second Limitation: "computing, a first subset of the segment and a second subset of the segment, wherein the first subset includes users of the segment that interact with the website a greater number of times than a threshold number of times during a range of the time series data and the second subset comprises users of the segment that interact with the website a lesser number of times than the threshold number of times during the range of the time series data"
Third Limitation: "generating, a segment return value by computing a first return value for the first subset and a second return value for the second subset, respectively, and combining the first return value and the second return value to obtain the segment return value, wherein the first return value comprises a numerical representation of users in the first subset who are predicted to interact with the website during a future time period subsequent to the first time period and the second return value comprises a numerical representation of users in the second subset who are predicted to interact with the website during the future time period"
Abstract Idea Grouping Analysis
Mental Processes (MPEP 2106.04(a)(2), Subsection III)
Limitations under their broadest reasonable interpretation, cover performance of the recited steps in the human mind, including observation, evaluation, and judgment. See MPEP 2106.04(a)(2), Subsection III.
The first limitation recites identifying a segment of users for a first time period based on time series data. Under its broadest reasonable interpretation, this step encompasses examining a collection of user interaction records and grouping users into a segment based on shared characteristics and temporal data. A human analyst provided with a set of interaction records for users visiting a website during a defined time period could examine those records and group users who share a relevant attribute such as browser type, device type, or location into a common segment using pen and paper. Compare Electric Power Group v. Alstom, (data analysis steps recited at a high level of generality remain mentally performable).
The second limitation recites computing first and second subsets of the segment by comparing each user's interaction count against a threshold number and partitioning users into one of two groups. Under its broadest reasonable interpretation, this step encompasses examining each user's interaction records, tallying that user's interactions, comparing the tally to a threshold number, and assigning the user to either the first or second subset accordingly. This is a binary classification based on a numerical comparison precisely the type of observation, evaluation, and judgment that a human can perform with pen and paper when provided the relevant records. See CyberSource Corp. v. Retail Decisions, Inc., (constructing a map of transactions based on account data was mentally performable); In re Killian, (collecting information and making determinations from multiple data sources is a mental process). The "prediction component" is a generic computer component whose nominal recitation does not alter this analysis.
The third limitation recites generating a segment return value by computing a first return value for the first subset and a second return value for the second subset each a numerical representation of predicted future returning users and combining those return values into a segment return value. Under its broadest reasonable interpretation, this step encompasses calculating a predicted count of returning users for each of two subsets using historical interaction data and summing those counts. A human analyst with access to historical interaction records could calculate return rates for each subset using pen and paper arithmetic and combine those figures into an aggregate prediction, even if such a task would be burdensome at scale. The use of a physical aid does not negate the mental nature of a limitation. See MPEP 2106.04(a)(2), Subsection III(B); October 2019 PEG Appendix 1, Example at 4 ("even if most humans would use a physical aid, like pen and paper or a calculator, to make such calculations, the use of a physical aid would not negate the mental nature of this limitation"). The step is recited at a high level of generality without specifying any particular algorithm, formula, or machine-dependent mechanism for computing the return values, covering any means of computing such numerical predictions including mental or pen-and-paper methods.
Mathematical Concepts (MPEP 2106.04(a)(2), Subsection I)
The limitations independently fall within the mathematical concepts grouping. The limitation recites comparing interaction counts against a threshold a mathematical comparison. Limitations recite computing numerical representations of predicted returning users for each of two subsets and combining those numerical representations a mathematical calculation. The specification itself describes these operations using explicit mathematical equations, including ratio calculations of the form, moving average estimators, ARMA models, ARIMA models, and linear regression formulas (see Specification pars. [0100]–[0131]). Computing numerical representations of users predicted to return using these mathematical methods constitutes a mathematical calculation within the mathematical concepts grouping. See SAP America, Inc. v. InvestPic, LLC, (resampled statistical analysis to generate a distribution is a mathematical calculation); Parker v. Flook, (computations representing predictions are mathematical calculations).
Because the limitations together recite both mental processes and mathematical concepts, and because there are no bright lines between the types of judicial exceptions, the Examiner identifies limitations together as constituting a single abstract idea falling within the mental processes and mathematical concepts groupings of abstract ideas, as instructed by MPEP 2106.04, Subsection II.B. See also 2024 AI-SME Update Examples at 23 (identifying claim limitations as reciting both a mathematical concept and a mental process and considering the limitations together as a single abstract idea for further analysis). Accordingly, the claims recite a judicial exception, and the analysis proceeds to Step 2A, Prong Two.
STEP 2A, Prong Two
Identification of Additional Elements
The following elements are identified as additional elements beyond the recited abstract idea:
• A segmentation component (generic computer component)
• A prediction component (generic computer component)
• A monitoring component (recited in dependent claims, generic computer component)
• A processor and memory component (claim 20, generic computer components)
• A non-transitory computer-readable medium (claim 17, generic computer storage)
• "Providing computing resources for the website based on the segment return value, wherein the computing resources comprise a bandwidth resource, a server resource, or a cloud-based service resource allocated for use by the website during the future time period when the segment return value indicates the segment of users will have a repeat interaction with the website during the future time period"
Analysis of Additional Elements
Mere Instructions to Apply the Exception (MPEP 2106.05(f))
The additional elements amount to no more than mere instructions to implement the abstract idea on generic computer components. The "segmentation component," "prediction component," and "monitoring component" are each recited as tools for performing the abstract steps of segmenting users, computing subsets, and generating return values they are generic computer components directed to carry out the abstract idea without any specific technical implementation detail. The "providing computing resources" limitation is a functional instruction to apply the result of the abstract computation to a resource allocation decision tantamount to adding the words "apply it" to the judicial exception by directing that the abstract prediction be used to allocate generic computing resources. See Alice Corp.; OIP Technologies, Inc. v. Amazon.com, Inc.; Internet Patents Corp. v. Active Network, Inc., (a limitation describing the effect or result dissociated from any method by which the result is accomplished does not provide a meaningful limitation). Examiners are reminded, however, not to oversimplify the claims or expand the "apply it" consideration. The Examiner has carefully evaluated the claim as a whole and as an ordered combination per McRO and Ex Parte Desjardins and has determined that even when all limitations are considered together, the claims do not recite a particular solution to a technological problem but rather recite the idea of applying an abstract prediction result to a generic resource allocation decision.
Insignificant Extra-Solution Activity (MPEP 2106.05(g))
The "providing computing resources" limitation constitutes insignificant post-solution activity appended to the abstract computational steps. It is the downstream application of the segment return value to a resource allocation decision and simply makes use of the computed output after the value is first determined. See Parker v. Flook, ("Post-solution activity, no matter how conventional or obvious in itself, can [not] transform an unpatentable principle into a patentable process."); Ultramercial, Inc. v. Hulu, LLC. The limitation recites no specific technical interaction between the prediction component and the resource provisioning system, no particular mechanism by which the segment return value technically controls the allocation, and no structural or architectural constraint on how the allocation is implemented. The Examiner will re-evaluate this element at Step 2B as instructed by MPEP 2106.05, Subsection I.A, and the 2019 PEG Introductory Module.
Field-of-Use / General Technological Environment (MPEP 2106.05(h))
To the extent not already addressed, the recitations of a "website," "web browser," "content channel," and the allocation of bandwidth, server, or cloud-based service resources to a website serve to generally link the abstract idea to the technological environment of web-based computing and website resource management without imposing any meaningful technical constraint on the abstract computation. These recitations identify the field in which the abstract idea is applied but do not change the nature of the abstract computation or provide any specific technical implementation that would integrate the judicial exception into a practical application. See Diamond v. Diehr, 4; MPEP 2106.05(h).
Step 2A, Prong Two Conclusion
Considering the additional elements individually and in combination, and evaluating the claim as a whole as an ordered combination without ignoring the requirements of the individual steps per McRO and Ex Parte Desjardins, the claims do not integrate the recited judicial exception into a practical application. The additional elements do not reflect an improvement to the functioning of a computer or to any other technology or technical field; do not constitute use of a particular machine; amount to mere instructions to apply the abstract idea using generic computer components; and constitute insignificant post-solution activity and field-of-use limitations. The claims are therefore directed to an abstract idea. The analysis proceeds to Step 2B.
STEP 2B
The additional elements identified in Step 2A, Prong Two are carried forward. The conclusions from Step 2A, Prong Two regarding the improvements consideration (MPEP 2106.05(a)), particular machine consideration (MPEP 2106.05(b)), and mere instructions to apply the exception consideration (MPEP 2106.05(f)) are also carried forward, as those conclusions are the same in Step 2B. See MPEP 2106.05, Subsection I.A.
Re-Evaluation of Insignificant Extra-Solution Activity
Upon re-evaluation of the "providing computing resources" limitation in Step 2B, the Examiner confirms that this element does not rise to the level of an unconventional or non-generic activity that would supply an inventive concept. As set forth in the Well-Understood, Routine, Conventional Activity analysis below, allocating bandwidth, server, and cloud-based service resources to a website in response to a computed output is well-understood, routine, and conventional activity in the field. Re-evaluation of this element therefore does not alter the Step 2B conclusion.
Well-Understood, Routine, Conventional Activity Analysis (MPEP 2106.05(d))
The additional elements, considered individually and in combination, are well-understood, routine, and conventional activities in the field, as supported by the following:
(A) Specification Evidence: The specification describes bandwidth, servers, and cloud-based services at paragraph [0027] in a single, general sentence without any technical elaboration regarding how resource allocation is accomplished: "An example of an efficient allocation of resources is determining an amount of bandwidth, servers, cloud-based services, and the like that should be provided for the content channel based on the segment return value." This description treats resource allocation as a self-evident, well-known downstream application of the segment return value without providing any technical detail regarding the allocation mechanism, thereby demonstrating that resource allocation using bandwidth, servers, and cloud-based services is sufficiently well-known that the specification need not describe the particulars to satisfy 35 U.S.C. 112(a). See MPEP 2106.05(d)(I)(A). Similarly, the specification describes at paragraph [0037]–[0038] that the apparatus communicates via cloud services, uses HTTP and SMTP protocols, and comprises a general-purpose computing device, a personal computer, a laptop, a mainframe, or a supercomputer all described as known, commercially available components. The specification's treatment of these elements confirms their well-understood, routine, and conventional nature in the field.
The Examiner additionally notes that the combination of the abstract computational steps with the "providing computing resources" limitation does not supply an inventive concept. An inventive concept cannot be furnished by the abstract idea itself. Genetic Technologies Ltd. v. Merial LLC; Alice Corp. The abstract computational steps identifying user segments, computing subsets based on interaction frequency, and generating a segment return value constitute the recited judicial exception and cannot supply the inventive concept. The only remaining additional element is the "providing computing resources" limitation, which, as established above, is well-understood, routine, and conventional activity.
The claims as a whole do not include additional elements that amount to significantly more than the recited judicial exception. The claims are not patent eligible.
DEPENDENT CLAIMS ANALYSIS
The dependent claims do not add limitations that integrate the judicial exception into a practical application or provide an inventive concept:
Claims 2–3 further define the first subset as including users who interact with the website more than one time during the first time period (claim 2), or at least one time during one or more time periods before the first time period (claim 3). These limitations further narrow the abstract mental process of classifying users based on interaction frequency by specifying the threshold value and the time range used in the classification. They add additional detail to the abstract idea itself without adding any additional element beyond the abstract idea that would integrate the judicial exception into a practical application or provide an inventive concept.
Claim 4 adds identifying a plurality of attributes characterizing the users and selecting an attribute on which the segment is based. These limitations further specify the abstract mental process of identifying and classifying users based on shared attributes. The selection of an attribute to define a segment is itself a mental observation, evaluation, and judgment that can be performed mentally or with pen and paper, and does not add any additional element that would provide an inventive concept.
Claim 5 adds generating a cookie for the user based on an interaction with the website, and determining that the user is in the segment based on the cookie. Generating a cookie and using it to track user identity is well-understood, routine, and conventional activity in web-based computing, as the specification itself describes at paragraph [0023] and [0095] without any technical elaboration, indicating that cookies are well-known identifying data. This limitation does not supply an inventive concept.
Claim 6 adds inserting monitoring code into the website and collecting time series data based on the code. Inserting monitoring code into a website and collecting data at a high level of generality thereby are well-understood, routine, and conventional data-gathering activities in web analytics, described in the specification at paragraph [0062] without technical elaboration. These are insignificant extra-solution data-gathering steps that do not supply an inventive concept.
Claim 7 adds generating customized content for the segment based on the segment return value. The generation of customized content based on a predicted user segment is itself a method of organizing human activity in the form of a commercial or marketing interaction. This limitation does not supply an inventive concept and may itself recite an additional abstract idea.
Claims 11-13 add computing a moving average estimator for the first return value based on a plurality of time periods before the first time period (claim 11), wherein the moving average estimator comprises an autoregressive moving average (claim 12), and computing a seasonal parameter of the time series data on which the moving average estimator is based (claim 13), These limitations add further mathematical detail to the abstract mathematical concept of computing the first return value using ARMA, ARIMA, or seasonal models. They add detail to the abstract idea itself, the mathematical computation, without adding any additional elements beyond the abstract idea. Mathematical computations, even complex ones, cannot supply an inventive concept. See SAP America.
Claim 15 adds that the segment return value comprises a ratio between the number of users predicted to interact during the future time period and the number of users in the segment during the first time period. This limitation adds a specific mathematical formula a ratio to the abstract mathematical computation, further specifying the abstract idea itself without adding any additional element beyond the abstract idea.
Claim 16 adds identifying a frequency for the time series data and selecting a model for computing the segment return value based on the frequency. These limitations add further abstract mathematical and mental process steps identifying a seasonal frequency and selecting a mathematical model based on that frequency that further specify the abstract idea without adding any additional element that would supply an inventive concept.
Claims 21–24 (dependent on claim 20) add a moving average estimator for the first return value (claim 21), a ratio segment return value (claim 22), identification and selection of user attributes for segmentation (claim 23), and identification of a data frequency and model selection (claim 24). For the same reasons as set forth with respect to claims 15, 4, and 16 above, respectively, these limitations either add mathematical detail to the abstract idea itself, add further abstract mental process steps, or add well-understood, routine, and conventional activity, and none supply an inventive concept.
For the foregoing reasons, claims 1–7, 11-13, and 15–24 are rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter. Should Applicant elect to amend the claims, the Examiner encourages Applicant to consider amendments that recite specific technical mechanisms, protocols, or architectural configurations by which computing resources are allocated in response to the segment return value, or that recite specific improvements to how the computing infrastructure itself operates that are not already described in the specification at a high level of generality.
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.
Claims 1-3, 5, 6, 15, 17, 20, 21, and 24 is/are rejected under 35 U.S.C. 103 as being unpatentable in view of Pattabiraman (U.S. Patent Publication 2021/0337032 A1) in view of Wu et al. (U.S. Patent Publication 2021/0150608 A1) (hereafter Wu) in further view of Aras et al. (U.S. Patent 5,884,037) (hereafter Aras).
Referring to Claim 1, Pattabiraman teaches a method for segment size estimation, said method comprising:
identifying, by a segmentation component, a segment of a plurality of users for a first time period based on time series data for a website, wherein the time series data includes a series of interactions between the plurality of users and the website via a web and wherein the segment includes a portion of the plurality of users interacting with the website during the first time period (see; par. [0029] of Pattabiraman teaches network utilization on a website interaction for a period of the user session, par. [0042] based on a user length of time of the user session).
Pattabiraman teaches network utilization on a website interaction for a period of the user session, based on a user length of time of the user session. However Pattabiraman does not explicitly disclose the following limitation, however,
Wu teaches computing, by a prediction component, a first subset of the segment and a second subset of the segment, wherein the first subset includes users of the segment that interact with the website a greater number of times than a threshold number of times during a range of the time series data and the second subset comprises users of the segment that interact with the website a lesser number of times than the threshold number of times during the range of the time series data (see: par. [0083] of Wu teaches determining...a first subset of users from the plurality of users, the first subset of users comprising users that have previously used a portable device in at least one second region to initiate a transaction at least a predetermined number of times... determining... a second subset of users from the plurality of users, the second subset of users comprising users that have not previously used a portable device to initiate a transaction in the at least one second region or have used a portable device to initiate a transaction in the at least one second region less than a predetermined number of times, par. [0084]... users in the first subset of users have used their portable financial device to initiate a transaction in the second region at least a predetermined number of times. This predetermined number of times may be a number of times determined by the transaction service provider 102...such that certain transaction parameters... provide a reliable prediction of that user's propensity to use their portable financial device in the second region in the future, the second subset of users may include users that have not used their portable financial device at all or less than a predetermined number of times in the past... in the second region to initiate a transaction),
generating, by the prediction component, a segment return value by computing a first return value for the first subset and a second return value for the second subset, respectively, and combining the first return value and the second return value to obtain the segment return value (see; par. [0105] of Wu teaches generating... an activation metric for each user of the first subset of users based at least partially on a first algorithm, a first plurality of transaction parameters, and transaction data for portable device transactions initiated by each user; generating... an activation metric for each user of the second subset of users based at least partially on a second algorithm... determining... a plurality of target users from the plurality of users based at least partially on the activation metrics...' (combining both subsets' metrics), par. [0106]-[0107] The activation metric may be generated for each user in the first subset by considering the first algorithm, the relevant transaction parameters, and the transaction data for each user to generate an activation metric for each user in the first subset, At a seventh step... the transaction service provider processor 112 generates an activation metric for each user of the second subset of users based at least partially on a second algorithm... The first subset and the second subset are combined after the activation metric is applied to yield the target users, and the plurality of target users may include the top 30% of all users. In other embodiments, the first subset and the second subset are separately considered, and the top 30% of users from the first subset and the top 30% of users from the second subset may constitute the plurality of target users. (i.e. (1) computing a first activation metric (first return value) for the first subset using a first algorithm); (2) computing a second activation metric (second return value) for the second subset using a second different algorithm); (3) combining the metrics from both subsets to identify target users ('the first subset and the second subset are combined after the activation metric is applied'). Under BRI, 'activation metric representing propensity for future activity' is the 'return value; combining subsets' metrics = combining return values to obtain segment return value), and
wherein the first return value comprises a numerical representation of users in the first subset who are predicted to interact with the website during a future time period subsequent to the first time period and the second return value comprises a numerical representation of users in the second subset who are predicted to interact with the website during the future time period (see; par. [0087] of Wu teaches the first activation metric represents a determined propensity of a user from the first subset of users to initiate future transaction activity in the at least one second region; the second activation metric represents a determined propensity of a user from the second subset of users to initiate future transaction activity in the at least one second region... par. [0087] (first subset metric): The activation metric may be generated for each user of the first subset of users based at least in part on a first algorithm, the first plurality of transaction parameters, and transaction data for portable financial device transactions initiated by each user, par. [0089] The activation metric may be generated for each user of the first subset of users based at least in part on a first algorithm, the first plurality of transaction parameters, and transaction data for portable financial device transactions initiated by each user, par. [0113] (second subset metric): 'The activation metric may be generated for each user of the second subset of users based at least in part on a second algorithm, the second plurality of transaction parameters, and transaction data for portable financial device transactions initiated by each user. FIGS. 10A-10B illustrate test data showing...how performing the previously-described processes in an activation model leads to more efficient target actions… high propensity cardholders in the second subset...were 3.4 times more likely to initiate transactions in the second region in the future. (Propensity metric = numerical representation of future return rate for that subset)).
The Examiner notes that Pattabiraman teaches similar to the instant application teaches optimizing network utilization based on segmenting users of online resources. Specifically, Pattabiraman discloses determining user sessions including number of times and actual amount of time and creating segmentation and analyzed data it is therefore viewed as analogous art in the same field of endeavor. Additionally, Wu teaches system, method, and computer program product for segmenting users in a region based on predicted activity and as it is comparable in certain respects to Pattabiraman which optimizing network utilization based on segmenting users of online resources and method as well as the instant application it is viewed as analogous art and is viewed as reasonably pertinent to the problem faced by the inventor. This provides support that it would be obvious to combine the references to provide an obviousness rejection.
It would be obvious to one of ordinary skill in the art to include in the task management
(system/method/apparatus) of Pattabiraman computing, by a prediction component, a first subset of the segment and a second subset of the segment, and combining the first return value and the second return value to obtain the segment return value, as taught by Wu since the claimed invention is merely a combination of old elements, and in the combination each element merely would have performed the same function as it did separately, and one of ordinary skill in the art would have recognized that the results of the combination were predictable. Both Pattabiraman and Wu share the common goal of segmenting user populations based on behavioral data to optimize the targeted delivery of content and benefits with reasonable expectation of success, as the combination requires only the straightforward application of a known and proven technique to an analogous behavioral dataset, and they do not contradict or diminish the other alone or combined.
Pattabiraman in view of Wu discloses network utilization on a website interaction for a period of the user session, based on a user length of time of the user session, However Pattabiraman in view of Wu fails disclose the following limitation, however,
Aras teaches providing computing resources for the website based on the segment return value, wherein the computing resources comprise a bandwidth resource, a server resource, or a cloud-based resource allocated for use by the website during the future time period when the segment return value indicates the segment of users will have a repeat interaction with the website during the future time period (see: col. 3, lines (1-23) of Aras teaches the present invention uses Seasonal Autoregressive Integrated Moving Average ("ARIMA") trend analysis to enhance reservation-based management systems... The NCBP uses a seasonal ARIMA model with coefficients generated by the SMGU to predict the amount of bandwidth at a link or system level used by elements that do not conform to the reservation policy... The predicted available bandwidth for a period of time is then utilized by the connection control process for allocating network resources to requesting computer systems, which is viewed to expressly teach using seasonal ARIMA to predict future bandwidth (network resource) usage from time series data and then allocating network bandwidth resources based on those predictions).
Discussion of Pattabiraman’s similarity to the instant application and Wu being comparable to Pattabiraman is found above. Additionally, Aras teaches method and system for displaying anomalies in time series data and as it is comparable in certain respects to Pattabiraman and Wu which optimizing network utilization based on segmenting users of online resources and method as well as the instant application it is viewed as analogous art and is viewed as reasonably pertinent to the problem faced by the inventor. This provides support that it would be obvious to combine the references to provide an obviousness rejection.
It would be obvious to one of ordinary skill in the art to include in the task management
(system/method/apparatus) of Pattabiraman and Wu providing computing resources for the website based on the segment return value, as taught by Aras since the claimed invention is merely a combination of old elements, and in the combination each element merely would have performed the same function as it did separately, and one of ordinary skill in the art would have recognized that the results of the combination were predictable. Additionally, the combination of Pattabiraman, Wu, and Aras teach sharing the common goal of segmenting user populations based on behavioral data to optimize the targeted delivery of content and benefits with reasonable expectation of success, as the combination requires only the straightforward application of a known and proven technique to an analogous behavioral dataset with the added benefit of a mechanism for forecasting future network capacity and they do not contradict or diminish the other alone or when combined.
Referring to Claim 2, see discussion of claim 1 above, while Pattabiraman in view of Wu in further view of Aras teaches the method above, Pattabiraman further discloses a method having the limitations of,
the first subset includes users that interact with the website more than one time during the first time period (see; par. [0093]-[0095] of Pattabiraman teaches an example of counting the number of sessions on the website on different devices during a time period).
Referring to Claim 3, see discussion of claim 1 above, while Pattabiraman in view of Wu in further view of Aras teaches the method above, Pattabiraman further discloses a method having the limitations of,
the first subset includes users that interact with the website at least one time during one or more time periods before the first time period (see; par. [0079] of Pattabiraman teaches noting a time stamp prior to a different time stamp related to a different time period).
Referring to Claim 5, see discussion of claim 1 above, while Pattabiraman in view of Wu in further view of Aras teaches the method above, Pattabiraman further discloses a method having the limitations of:
generating, by a monitoring component, a cookie for the user based on an interaction of the user with the website (see; par. [0069] of Pattabiraman teaches monitoring users utilizing cookies to monitor the user interactions on the website).
determining, by the segmentation component, that the user is in the segment based on the cookie, (see; par. [0069]-[0071] of Pattabiraman teaches using cookies which is a user identifier that is used to determine segments, par. [0008] to provide personalized content).
Referring to Claim 6, see discussion of claim 1 above, while Pattabiraman in view of Wu in further view of Aras teaches the method above, Pattabiraman does not explicitly discloses a method having the limitations of,
inserting, by a monitoring component, code for monitoring the website in the website (see; Abstract of Pattabiraman teaches code that provides, par. [0042] a website to monitor user sessions).
collecting, by the monitoring component, the time series data based on the code (see; par. [0069]-[0070] of Pattabiraman teaches collecting time stamps associated with the data (i.e. time series)).
Referring to Claim 15, see discussion of claim 1 above, while Pattabiraman in view of Wu in further view of Aras teaches the method above, Pattabiraman further discloses a method having the limitations of:
the segment return value comprises a ratio between a number of users predicted to interact with the content channel during the second time period and a number of users in the segment during the first time period (see; par. [0018]-[0019] of Pattabiraman teaches segmentation based on a predicted outcomes and improve the prediction
Referring to Claim 17, Pattabiraman in view of Wu in further view of Aras teaches a non-transitory computer readable medium storing code. Claim 17 recites the same or similar limitations as those addressed above in claim 1, Claim 17 is therefore rejected for the same reasons as set forth above in claim 1, except for the following noted exceptions,
At least one processor, cause the at least one processor to perform operations (see; par. [0098] of Pattabiraman teaches a processor).
computing, by a prediction component, a first subset of a segment of the plurality of users and a second subset of the segment of the plurality of users based on the time series data (see: par. [0018]-[0019] of Pattabiraman teaches predict segmentation based on the amount of time a user spent during the session, par. [0093] taking into account the number of user sessions, par. [0016] where a website is utilized that indicates less than a threshold time of user sessions, par.[0093] taking into account number of sessions during a period of time (i.e. time series)).
Referring to Claim 20, Pattabiraman in view of Wu in further view of Aras teaches a system. Claim 20 recites the same or similar limitations as those addressed above in claim 1, and 17. Claim 20 is therefore rejected for the same reasons as set forth above in claim 1 and 17, except for the following noted exception,
a processor, a memory including instructions executable by the processor (see; col. 15, lines (9-11) a process computer that executes a program.
Referring to Claim 21, see discussion of claim 20 above, while Pattabiraman in view of Wu in further view of Aras teaches the system above, specifically Pattabiraman teaches network utilization on a website interaction for a period of the user session, based on a user length of time of the user session. However Pattabiraman fails to disclose the following limitation, however,
Wu teaches computing for the first return value based on a plurality of time periods before the first time period, wherein the first return value is based on the moving average estimator (see; par. [0087]-[0090] of Wu teaches a predictive model using multiple prior time periods for first-subset metric, which is viewed as computing the first-subset activation metric (first return value) using a predictive model over multiple prior time periods).
Pattabiraman teaches optimizing network utilization based on segmenting users of online resources and Wu teaches segmenting users in a region based on predicted activity. It would be obvious to one of ordinary skill in the art to include in the task management
(system/method/apparatus) of Pattabiraman computing for the first return value based on a plurality of time periods before the first time period, wherein the first return value is based on the moving average estimator as taught by Wu since the claimed invention is merely a combination of old elements, and in the combination each element merely would have performed the same function as it did separately, and one of ordinary skill in the art would have recognized that the results of the combination were predictable. Pattabiraman and Wu share the common goal of segmenting user populations based on behavioral data to optimize the targeted delivery of content and benefits with reasonable expectation of success, as the combination requires only the straightforward application of a known and proven technique to an analogous behavioral dataset, and they do not contradict or diminish the other alone or combined.
Pattabiraman in view of Wu discloses network utilization on a website interaction for a period of the user session, based on a user length of time of the user session, However Pattabiraman in view of Wu fails disclose the following limitation, however,
Aras teaches a moving average estimator (see; col. 3, lines (32-45) of Aras teaches ARIMA moving average terms based on multiple prior time period lags, which is viewed as a moving average estimator (ARIMA MA terms) based on multiple past time-period observations).
Discussion of Pattabiraman’s similarity to the instant application and Wu being comparable to Pattabiraman is found above. Additionally, Aras teaches method and system for displaying anomalies in time series data and as it is comparable in certain respects to Pattabiraman and Wu which optimizing network utilization based on segmenting users of online resources and method as well as the instant application it is viewed as analogous art and is viewed as reasonably pertinent to the problem faced by the inventor. This provides support that it would be obvious to combine the references to provide an obviousness rejection.
It would be obvious to one of ordinary skill in the art to include in the task management
(system/method/apparatus) of Pattabiraman and Wu a moving average estimator as taught by Aras since the claimed invention is merely a combination of old elements, and in the combination each element merely would have performed the same function as it did separately, and one of ordinary skill in the art would have recognized that the results of the combination were predictable. Additionally, the combination of Pattabiraman, Wu, and Aras teach sharing the common goal of segmenting user populations based on behavioral data to optimize the targeted delivery of content and benefits with reasonable expectation of success, as the combination requires only the straightforward application of a known and proven technique to an analogous behavioral dataset with the added benefit of a mechanism for forecasting future network capacity and they do not contradict or diminish the other alone or when combined.
Referring to Claim 24, see discussion of claim 20 above, while Pattabiraman in view of Wu in further view of Aras teaches the system above, Pattabiraman and Wu discloses determining user sessions including number of times and actual amount of time and creating segmentation and analyzed data. However, Pattabiraman and Wu fails to disclose the following, however,
Aras teaches identifying a frequency for the time series data; and selecting a model for computing the segment return value based on the frequency (see; col. 3, lines (1-14), and col. 5, lines (1-5) of Aras teaches dentifying time series periodicity/frequency and selecting/parameterizing the ARIMA model based on that frequency).
Aras teaches method and system for displaying anomalies in time series data and as it is comparable in certain respects to Pattabiraman and Wu which optimizing network utilization based on segmenting users of online resources. It would be obvious to one of ordinary skill in the art to include in the task management (system/method/apparatus) of Pattabiraman and Wu identifying a frequency for the time series data; and selecting a model for computing the segment return value based on the frequency as taught by Aras since the claimed invention is merely a combination of old elements, and in the combination each element merely would have performed the same function as it did separately, and one of ordinary skill in the art would have recognized that the results of the combination were predictable. Additionally, the combination of Pattabiraman, Wu, and Aras teach sharing the common goal of segmenting user populations based on behavioral data to optimize the targeted delivery of content and benefits with reasonable expectation of success, as the combination requires only the straightforward application of a known and proven technique to an analogous behavioral dataset with the added benefit of a mechanism for forecasting future network capacity and they do not contradict or diminish the other alone or when combined.
Claims 4, 7, 15, 16, and 19 is/are rejected under 35 U.S.C. 103 as being unpatentable in view of Pattabiraman (U.S. Patent Publication 2021/0337032 A1) in view of Wu et al. (U.S. Patent Publication 2021/0150608 A1) (hereafter Wu) in further view of Aras et al. (U.S. Patent 5,884,037) (hereafter Aras). in further view of Storan et al. (U.S. Patent Publication 2010/0269134 A1) (hereafter Storan).
Referring to Claim 4, see discussion of claim 1 above, while Pattabiraman in view of Wu in further view of Aras teaches the method above, Pattabiraman in view of Wu in further view of Aras does not explicitly disclose a method having the limitations of, however,
Storan teaches identifying, by the segmentation component, a plurality of attributes characterizing the plurality of users (see; Abstract of Storan teaches determining a segment based on a user profile data),
selecting, by the segmentation component, an attribute of the plurality of attributes, wherein the segment is identified based on the selected attribute (see; par. [0012] of Storan teaches proving content based on determined attributes).
Discussion of Pattabiraman’s similarity to the instant application with Wu and Aras being comparable to Pattabiraman is found above. Additionally, Storan teaches television programing promotion and as it is comparable in certain respects to Pattabiraman, Wu, and Aras which optimizing network utilization based on segmenting users of online resources and method as well as the instant application it is viewed as analogous art and is viewed as reasonably pertinent to the problem faced by the inventor. This provides support that it would be obvious to combine the references to provide an obviousness rejection.
It would be obvious to one of ordinary skill in the art to include in the task management
(system/method/apparatus) of Pattabiraman, Wu and Aras identifying, by the segmentation component, a plurality of attributes characterizing the plurality of users and selecting, by the segmentation component, an attribute of the plurality of attributes, wherein the segment is identified based on the selected attribute as taught by Storan since the claimed invention is merely a combination of old elements, and in the combination each element merely would have performed the same function as it did separately, and one of ordinary skill in the art would have recognized that the results of the combination were predictable. Additionally, the combination of Pattabiraman, Wu, Aras, and Storan teach sharing the common goal of segmenting user populations based on behavioral data to optimize the targeted delivery of content and benefits with reasonable expectation of success, as the combination requires only the straightforward application of a known and proven technique to an analogous behavioral dataset with the added benefit of a mechanism for forecasting future network capacity and inputs to its modal segmentation engine for enhanced precision and they do not contradict or diminish the other alone or when combined.
Referring to Claim 7, see discussion of claim 1 above, while Pattabiraman in view of Wu in further view of Aras teaches the method above, Pattabiraman in view of Wu in further view of Aras does not explicitly disclose a method having the limitations of, however,
Storan teaches generating, by the content component, the customized content for the segment based on the segment return value (see; par. [0008] of Storan teaches determining optimal time and channels to deliver content to a specific user (i.e. customized content), Abstract based on a determined segment).
Discussion of Pattabiraman’s similarity to the instant application with Wu and Aras being comparable to Pattabiraman is found above. Additionally, Storan teaches television programing promotion and as it is comparable in certain respects to Pattabiraman, Wu, and Aras which optimizing network utilization based on segmenting users of online resources and method as well as the instant application it is viewed as analogous art and is viewed as reasonably pertinent to the problem faced by the inventor. This provides support that it would be obvious to combine the references to provide an obviousness rejection.
It would be obvious to one of ordinary skill in the art to include in the task management
(system/method/apparatus) of Pattabiraman, Wu and Aras generating, by the content component, the customized content for the segment based on the segment return value as taught by Storan since the claimed invention is merely a combination of old elements, and in the combination each element merely would have performed the same function as it did separately, and one of ordinary skill in the art would have recognized that the results of the combination were predictable. Additionally, the combination of Pattabiraman, Wu, Aras, and Storan teach sharing the common goal of segmenting user populations based on behavioral data to optimize the targeted delivery of content and benefits with reasonable expectation of success, as the combination requires only the straightforward application of a known and proven technique to an analogous behavioral dataset with the added benefit of a mechanism for forecasting future network capacity and inputs to its modal segmentation engine for enhanced precision and they do not contradict or diminish the other alone or when combined.
Referring to Claim 16, see discussion of claim 1 above, while Pattabiraman in view of Wu in further view of Aras teaches the method above, Pattabiraman in view of Wu in further view of Aras does not explicitly disclose a method having the limitations of, however,
Storan teaches identifying, by the prediction component, a frequency for the time series data (see; par. [0041] of Storan teaches determining a frequency of a user watches and the changes when they watch during different time periods), and
selecting, by the prediction component, a model for computing the segment return value based on the frequency (see; par. [0085]-[0086] of Storan teaches a model that projects a segment based on analyzed data, par. [0041] based on determining a frequency of a user watches and the changes when they watch during different time periods).
Discussion of Pattabiraman’s similarity to the instant application with Wu and Aras being comparable to Pattabiraman is found above. Additionally, Storan teaches television programing promotion and as it is comparable in certain respects to Pattabiraman, Wu, and Aras which optimizing network utilization based on segmenting users of online resources and method as well as the instant application it is viewed as analogous art and is viewed as reasonably pertinent to the problem faced by the inventor. This provides support that it would be obvious to combine the references to provide an obviousness rejection.
It would be obvious to one of ordinary skill in the art to include in the task management
(system/method/apparatus) of Pattabiraman, Wu and Aras identifying, by the prediction component, a frequency for the time series data, selecting, by the prediction component, a model for computing the segment return value based on the frequency as taught by Storan since the claimed invention is merely a combination of old elements, and in the combination each element merely would have performed the same function as it did separately, and one of ordinary skill in the art would have recognized that the results of the combination were predictable. Additionally, the combination of Pattabiraman, Wu, Aras, and Storan teach sharing the common goal of segmenting user populations based on behavioral data to optimize the targeted delivery of content and benefits with reasonable expectation of success, as the combination requires only the straightforward application of a known and proven technique to an analogous behavioral dataset with the added benefit of a mechanism for forecasting future network capacity and inputs to its modal segmentation engine for enhanced precision and they do not contradict or diminish the other alone or when combined.
Referring to Claim 19, see discussion of claim 17 above, while Pattabiraman in view of Wu in further view of Aras teaches the method above Claim 19 recites the same or similar limitations as those addressed above in claim 15, Claim 19 is therefore rejected for the same or similar limitations as set forth above in claim 15.
Claim 22 and 23 is/are rejected under 35 U.S.C. 103 as being unpatentable in view of Pattabiraman (U.S. Patent Publication 2021/0337032 A1) in view of Wu et al. (U.S. Patent Publication 2021/0150608 A1) (hereafter Wu) in further view of Aras et al. (U.S. Patent 5,884,037) (hereafter Aras). in further view of Weinstein et al. (U.S. Patent Publication 2014/0089043 A1) (hereafter Weinstein).
Referring to Claim 22, see discussion of claim 20 above, while Pattabiraman in view of Wu in further view of Aras teaches the system above, specifically Pattabiraman discloses determining user sessions including number of times and actual amount of time and creating segmentation and analyzed data, However, Pattabiraman in view of Wu in further view of Aras fails to disclose.
Weinstein teaches the segment return value comprises a ratio between a number of users predicted to interact with the website during the future time period and a number of users in the segment during the first time period (see; Par. [0075] and par. [0072] of Weinstein teaches (future size / current size ratio), which is viewed as a future/current segment size ratio at a designated time period).
Discussion of Pattabiraman’s similarity to the instant application with Wu and Aras being comparable to Pattabiraman is found above. Additionally, Weinstein teaches audience size estimation and complex segment logic and as it is comparable in certain respects to Pattabiraman, Wu, and Aras which optimizing network utilization based on segmenting users of online resources and method as well as the instant application it is viewed as analogous art and is viewed as reasonably pertinent to the problem faced by the inventor. This provides support that it would be obvious to combine the references to provide an obviousness rejection.
It would be obvious to one of ordinary skill in the art to include in the task management
(system/method/apparatus) of Pattabiraman, Wu, and Aras the segment return value comprises a ratio between a number of users predicted to interact with the website during the future time period and a number of users in the segment during the first time period as taught by Weinstein since the claimed invention is merely a combination of old elements, and in the combination each element merely would have performed the same function as it did separately, and one of ordinary skill in the art would have recognized that the results of the combination were predictable. Additionally, the combination of Pattabiraman, Wu, Aras, and Weinstein teach sharing the common goal of segmenting user populations based on behavioral data to optimize the targeted delivery of content and benefits with reasonable expectation of success, as the combination requires only the straightforward application of a known and proven technique to an analogous behavioral dataset with the added benefit of a mechanism for forecasting future network capacity and real time segment sizing and complex rule evaluations and they do not contradict or diminish the other alone or when combined.
Referring to Claim 23, see discussion of claim 20 above, while Pattabiraman in view of Wu in further view of Aras teaches the system above, specifically Pattabiraman discloses determining user sessions including number of times and actual amount of time and creating segmentation and analyzed data, However, Pattabiraman in view of Wu in further view of Aras fails to disclose.
Weinstein teaches identifying a plurality of attributes characterizing the plurality of users; and selecting an attribute of the plurality of attributes, wherein the segment is identified based on the selected attribute (see; par. [0034], [0041], and par. [0064] of Weinstein teaches plurality of user attributes; selecting a trait; segment identified from selected trait, which is viewed to teach identifying and selecting traits (attributes) to identify the segment).
Discussion of Pattabiraman’s similarity to the instant application with Wu and Aras being comparable to Pattabiraman is found above. Additionally, Weinstein teaches audience size estimation and complex segment logic and as it is comparable in certain respects to Pattabiraman, Wu, and Aras which optimizing network utilization based on segmenting users of online resources and method as well as the instant application it is viewed as analogous art and is viewed as reasonably pertinent to the problem faced by the inventor. This provides support that it would be obvious to combine the references to provide an obviousness rejection.
It would be obvious to one of ordinary skill in the art to include in the task management
(system/method/apparatus) of Pattabiraman, Wu, and Aras identifying a plurality of attributes characterizing the plurality of users; and selecting an attribute of the plurality of attributes, wherein the segment is identified based on the selected attribute as taught by Weinstein since the claimed invention is merely a combination of old elements, and in the combination each element merely would have performed the same function as it did separately, and one of ordinary skill in the art would have recognized that the results of the combination were predictable. Additionally, the combination of Pattabiraman, Wu, Aras, and Weinstein teach sharing the common goal of segmenting user populations based on behavioral data to optimize the targeted delivery of content and benefits with reasonable expectation of success, as the combination requires only the straightforward application of a known and proven technique to an analogous behavioral dataset with the added benefit of a mechanism for forecasting future network capacity and real time segment sizing and complex rule evaluations and they do not contradict or diminish the other alone or when combined.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Rastaogi (U.S. Patent 9,430,535 B2) discloses a method and apparatus for normalizing and predicting time series data.
Alvarez Callau et al. (U.S. Patent 10,873,541 B2) discloses systems and methods for proactively and reactively allocating resources in cloud based networks.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to STEPHEN S SWARTZ whose telephone number is (571)270-7789. The examiner can normally be reached Mon-Fri 9:00 - 6:00.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Boswell Beth can be reached at 571 272-6737. 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.
/S.S.S/Examiner, Art Unit 3625
/BETH V BOSWELL/Supervisory Patent Examiner, Art Unit 3625
