DETAILED ACTION
Status
This communication is in response to Applicant’s “AMENDMENT AND REPLY UNDER 37 C.F.R. § 1.111” filed on March 9, 2026 (hereinafter “Amendment”). In the Amendment, Applicant amended Claims 1, 10 and 19; cancelled no claim(s); and added no claim(s). Claims 7 and 16 were previously cancelled. Therefore, Claims 1-6, 8-15 and 17-22 remain pending and presented for examination. Of the pending claims, Claims 1, 10 and 19 remain independent claims.
The present application, filed after March 16, 2013, is being examined under the first inventor to file (FITF) provisions of the America Invents Act (AIA ).
Examiner notes this case (i.e., U.S. Patent App. No. 17/960,618) has published as
U.S. Patent Application Publication No. 2024/0119484 of Kiran RAMA (hereinafter “Rama”).
Examiner notes Recentive Analytics, Inc. v. Fox Corp., Case No. 2023-2437 (Fed. Cir. Apr. 18, 2025), which was decided on April 18, 2025, by the US Court of Appeals for the Federal Circuit (CAFC).
Priority/Benefit Claim
No claim(s) for benefit or priority exists in this application and, therefore, the effective filing date of this application is its filing date of October 5, 2022.
CPC Classification Notes
Examiner notes CPC classifications G06Q 30/0251; G06Q 30/0269; and G06Q 30/0271 — each of which encompasses targeted-based advertising:
G06Q 30/00 Commerce
G06Q 30/02 • Marketing Price estimation or determination;
G06Q 30/0241 •• Advertisements
G06Q 30/0251 ••• Targeted advertisements
G06Q 30/0269 •••• {based on user profile or attribute}
G06Q 30/0271 ••••• {Personalized advertisement}
Information Disclosure Statement (IDS)
Examiner notes U.S. Patent Application No. 17/899,504 was mentioned in Applicant’s IDS filed on 12/012025 and has published as U.S. Patent Application Publication No. 2024/0070475 of HUANG et al. ( “Huang”).
Applicant is notified of MPEP § 2001.06(b): “prior art references from one application must be made of record in another subsequent application if such prior art references are ‘material to patentability’ of the subsequent application”.
Applicant is also notified of 37 C.F.R. 1.56, which states that each inventor named in the application has a duty to disclose information material to patentability.
Examiner Notes
Examiner notes the following instances of parallel language recited in Applicant’s claims:
Independent Claims 1, 10 and 19 recite parallel claims limitations to each other.
Claim 10 recites a method implemented by a computing system, the method comprising processes/steps recited in Claim 1.
Claims 11-12 recite parallel subject matter to that of respective Claims 2-3.
Claims 14-15 recite parallel subject matter to that of respective Claims 5-6.
Claims 17-18 recite parallel subject matter to that of respective Claims 8-9.
Claim 19 recites a computer program product comprising a computer-readable storage medium having instructions recorded thereon for enabling a processor-based system to perform operations/processes recited in Claim 1.
Claim 20 recites parallel subject matter to that of Claim 4.
Claim 21 recites parallel subject matter to that of Claim 8 (Claim 17 too).
Response to Amendments
A Summary of the Response to Applicant’s Amendment:
Applicant’s Amendment does not overcome rejections to Claims 1-6, 8-15 and 17-22 under 35 U.S.C. § 101; therefore, the Examiner maintains/submits § 101 rejections to Claims 1-6, 8-15 and 17-22, as provided below.
Applicant’s Amendment introduces new rejections to independent Claims 1, 10 and 19 under 35 U.S.C. § 112(b) of the AIA ; therefore, the Examiner asserts § 112(b) rejections to Claims 1-6, 8-15 and 17-22, as provided below.
Applicant’s arguments are found to be not persuasive; please see Examiner’s “Response to Arguments” provided below.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b) of the America Invents Act (AIA ):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
Claims 1-6, 8-15 and 17-22 are rejected under 35 U.S.C. 112(b) of the AIA as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention. “A claim is indefinite when it contains words or phrases whose meaning is unclear” (MPEP § 2173.05(e)).
Regarding independent Claims 1, 10 and 19, since it is unclear as to what the phrase “the action” (bolding emphasis added by Examiner) makes antecedent reference to in each of Applicant’s independent claims — there is insufficient antecedent basis for the phrase “the action” recited in each independent claim — independent Claims 1, 10 and 19 are rejected under 35 U.S.C. 112(b) of the AIA as being indefinite. For example, it is unclear as to whether the phrase “the action” recited in phrase “… as a result of the targeted entities satisfying the criterion, perform the action with regard to the targeted entities…” (bolding emphases added by Examiner):
refers to recited “a designated operation” (recited in each of Claims 1, 10 and 19);
refers to “approximating personalized targeting” (recited in each of Claims 1, 10 and 19);
refers to “using a trained algorithm of a machine learning model” (recited in each of Claims 1 and 10);
refers to “an action” that is not specifically/positively recited in Claims 1, 10 and 19; or
refers to some combination thereof (i.e., some combination of 1, 2, 3 and 4).
As currently presented, there is insufficient antecedent basis for “the action” recited in each of Applicant’s independent claims. In addition, it is unclear what the phrase “perform the action” means in the context of Applicant’s independent claims. Consequently, independent Claims 1, 10 and 19 are rejected under 35 U.S.C. 112(b) of the AIA as being indefinite. Appropriate corrections are required.
Claims 2-6, 8-9 and 22 each depend directly from independent Claim 1, but do not resolve the above issues and inherit the deficiencies of Claim 1; therefore, Claims 2-6, 8-9 and 22 are rejected under 35 U.S.C. 112(b) of the AIA .
Similarly, Claims 11-15 and 17-18 depend directly from independent Claim 10, but do not resolve the above issues and inherit the deficiencies of Claim 10; therefore, Claims 11-15 and 17-18 are rejected under 35 U.S.C. 112(b) of the AIA .
Similarly, Claims 20 and 21 depend directly from Claim 19, but do not resolve the above issues and inherits the deficiencies of independent Claim 19; therefore, Claims 20-21 are rejected under 35 U.S.C. 112(b) of the AIA .
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 1-6, 8-15 and 17-22 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. During patent examination, the pending claims must be “given their broadest reasonable interpretation consistent with the specification” (MPEP § 2111). In view of this standard and based upon consideration of all of the relevant factors with respect to each claim as a whole, Claims 1-6, 8-15 and 17-22 are rejected as ineligible subject matter under 35 U.S.C. 101.
Step 1: Claims 1-6, 8-15 and 17-22 satisfy Step 1 enunciated in Alice Corp. v. CLS Bank International, 573 U.S. __, 134 S. Ct. 2347 (2014) based on Applicant’s disclosure clarifying/defining that “A computer-readable storage medium is not a signal, such as a carrier signal or a propagating signal” (see specification paragraph [0099] on page 32 of Applicant’s originally-filed specification).
Step 2A: Claims 1-6, 8-15 and 17-22 are rejected under § 101 because Applicant’s claimed subject matter is directed to an abstract idea without significantly more. The rationale for this finding is that Applicant’s claims recite organizing/classifying/segmenting entities (e.g., users, purchasers, etc.) into one or more targetable categories (e.g., market segments or “bins”, such as Bin1, Bin2, Bin3 and Bin4 as illustrated in Figure 5 of Applicant’s drawings) based on behavioral information about the entities (i.e., recited “features that are associated with the entities” and likelihoods/probabilities of performing an action/operation, such as a “purchase” per ¶ [0036] of Applicant’s as-filed specification), and establishing a rule to target entities that have been classified within the one or more targetable categories (e.g., market segments) based on the behavioral information — segmenting a market into groups to be targeted (e.g., target market segmenting), such as for targeted advertising or contextual advertising —— (i.e., recited “approximating personalized targeting” such as, for example, “…triggering issuance of a discount coupon notification to each entity” that qualifies —— see Applicant’s originally-filed specification at paragraph [0050] on page 19, for instance), as more particularly recited in Applicant’s pending claims save for recited (non-abstract claim elements):
using a trained algorithm of a machine learning model (using a trained machine learning model);
providing inputs to the trained algorithm of the machine learning model;
performing the action;
(only Claim 1 and corresponding dependent claims) a system comprising non-transitory memory and a configured processing system coupled to the memory
(only Claim 10 preamble) a computing system; and
(only Claim 19 and corresponding dependent claim) a computer program product comprising a computer-readable storage medium having recorded instructions to enable a processor-based system to perform operations; and each of the recited steps/operations of “providing”.
Examiner notes Applicant’s abstract idea encompasses identifying/predicting user segments likely to perform a designated action/operation (e.g., clicking, viewing, interacting, etc.) toward a content item (e.g., advertisement, promotion, or “discount coupon notification” per Applicant’s as-filed specification at paragraph [0050] on page 19) by segmenting a group of users into classifications (e.g., market segmentation or classification, behavioral/classification segmentation, etc.). However, Examiner notes that utilizing behavioral information about entities to organize/classify/segment the entities into one or more targetable categories and to establish a targeting rule based on classifications of the entities within the targetable categories for “approximating personalized targeting” (e.g., targeted advertising), as currently recited in Applicant’s pending claims and further explained below, is within a certain method of organizing human activity — (i) fundamental economic principle or practice; and/or (ii) commercial interaction (including advertising, marketing or sales activities or behaviors; business relations). MPEP 2106.04(a)(2)(II)(A) provides examples of “fundamental economic principles or practices” and MPEP 2106.04(a)(2)(II)(B) provides additional discussion and examples of commercial or legal interactions. This judicial exception (i.e., abstract idea exception) is not integrated into a practical application because each claim as a whole, having the combination of additional elements beyond the judicial exception(s), does not integrate the exception into a practical application of the exception and, therefore, the pending claims are “directed to” a judicial exception under USPTO Step 2A. More specifically, each claim as a whole does not appear to reflect the combination of additional elements as: (1) improving the functioning of a computer itself or improving another technology or technical field, (2) applying the judicial exception with, or by use of, a particular machine/manufacture that is integral to the claim, (3) effecting a transformation or reduction of a particular article to a different state or thing, or (4) applying or using the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment. Instead, any improvement is to the underlying abstract idea of utilizing behavioral information about entities to organize/classify/segment the entities into one or more targetable categories and establishing a targeting rule based on classifications / segmentations of the entities within the targetable categories — target market segmenting for “approximating personalized targeting” (e.g., targeted advertising) as recited in each of Applicant’s independent claims. SAP Am., Inc. v. InvestPic, LLC, No. 2017-2081, 2018 U.S. App. LEXIS 12590, Slip. Op. 13 (Fed. Cir. May 15, 2018) (“What is needed is an inventive concept in the non-abstract realm.”). Examiner notes that each of Applicant’s independent Claims 1, 10 and 19 recites “using a trained algorithm of a machine learning model”; however, using or usage of “a trained algorithm of a machine learning model”, as recited in independent Claims 1, 10 and 19, appears as a high-level black box with no detail about the machine learning algorithm itself or any machine learning processes based on the recited “inputs”, such as how Applicant's model/tool operates on the recited “inputs” to produce an outputs (such as previously-recited “relationships”) and computed ranks. Furthermore, Examiner notes that no detail of any training algorithm appears to be mentioned in
Applicant's claims and, therefore, no specific way of training the algorithm/model exists within Applicant's recited use of a trained algorithm of a machine learning model. In addition, Examiner notes that the processes/operations of “using a trained algorithm of a machine learning model”, as recited in independent Claims 1, 10 and 19, “may be applied using any arbitrary machine learning model” — “model-agonistic” — per Applicant’s specification at paragraph [0056] (bolding emphases added by Examiner), which is similar to the analysis in Recentive Analytics, Inc. v. Fox Corp., Case No. 2023-2437 (Fed. Cir. Apr. 18, 2025), which found “The machine learning technology described…is conventional, as the…specifications demonstrate” (page 11 of Recentive Analytics, Inc. v. Fox Corp.). Also similar to Recentive Analytics, Inc. v. Fox Corp., Case No. 2023-2437 (Fed. Cir. Apr. 18, 2025), Applicant’s claims “do no more than claim the application of generic machine learning to new data environments, without disclosing improvements to the machine learning model[] to be applied” (see page 18 of Recentive Analytics, Inc. v. Fox Corp.). Applicant's mere recitations to “using a trained algorithm of a machine learning model” are not sufficient to amount to a practical application under Step 2A, Prong 2 of the Subject Matter Eligibility (SME) analysis in view of Recentive Analytics, Inc. v. Fox Corp., Case No. 2023-2437 (Fed. Cir. Apr. 18, 2025). In addition, although Applicant’s pending claims require computing ranks, categorizing, ordering, selecting, defining an interval, combining overlaps — please see Figure 5 of Applicant’s drawings in view of specification paragraphs [0058]–[0064] of Applicant’s original disclosure — these techniques encompass mathematical concepts in the form of formulas, equations, and calculations which also have been determined to constitute abstract ideas. See Memorandum, "Grouping of Abstract Ideas" and cases cited in footnote 12, such as enumerated in Section I of the 2019 Revised Patent Subject Matter Eligibility Guidance (84 Fed. Reg. 50). As noted on page 4 of the “October 2019 Update: Subject Matter Eligibility” issued by the USPTO, Examiner notes that a claim does not have to recite the word “calculating” in order to be considered a mathematical calculation. For example, a step of “determining” a variable or number using mathematical methods or “performing” a mathematical operation may also be considered mathematical calculations when the broadest reasonable interpretation (BRI) of the claim, in light of the specification, encompasses one or more mathematical calculations. Applicant’s additional elements, taken individually and in combination, do not appear to be integrated into a practical application since they embody mere instructions to implement the abstract idea on a computer or mere use of a computer as a tool to perform the abstract idea, do no more than generally linking the use of Applicant’s abstract idea to a particular technological environment or field of use {e.g., an online computer network of servers and user devices}, with independent Claims 1, 10 and 19 amounting to no more than combining the abstract idea with insignificant extra-solution activity including Applicant’s recited operations of “providing” and “performing the action” (Claims 1, 10 and 19), as further explained below. For the reasons discussed above, Applicant’s pending claims are directed to an abstract idea that is not integrated into a practical application under Step 2A, Prong 2 of the Subject Matter Eligibility (SME) analysis of 35 U.S.C. 101.
Step 2B: Under Step 2B enunciated in Alice Corp. v. CLS Bank International, 573 U.S. __, 134 S. Ct. 2347 (2014), Applicant’s instant claims do not recite limitations, taken individually and in combination, that are sufficient to amount to “significantly more” than the abstract idea because Applicant’s claims do not recite, as further explained in detail below, an improvement to another technology or technical field, an improvement to the functioning of a computer itself, an application with or by a particular machine, a transformation or reduction of a particular article to a different state or thing, unconventional steps confining the claim to a particular useful application, or meaningful limitations beyond generally linking the use of an abstract idea to a particular technological environment. Examiner notes that each of Claims 10-15 and 17-18 is drawn to a method; however, the method steps do not recite, require, or indicate implementation by a particular machine since none of limitations recited in Applicant’s method claims (Claims 10-15 and 17-18) are performed by any computer or processing device since recited method “implemented by a computing system” encompasses a situation where the computing system does no more than assist/help a person implement such steps/processes or thoughts when the person is using the computing system. Even if a computer/machine was implied, such as recited “a computing system” (Claim 10 preamble), Applicant’s limitations in Claims 10-15 and 17-18, taken individually and in combination, would be merely instructions to implement the abstract idea on a computer and would require no more than generally linking the use of an abstract idea to a particular technological environment or field of use (e.g., an online computer network of servers and user devices). Examiner also notes that albeit limitations recited in the Claims 1-6, 8-9 and 22 are performed by the generically recited “processing system coupled to…memory” while Claims 19-20 are performed by the generically recited “a processor-based system”, these claim limitations taken individually and in combination are merely instructions to implement the abstract idea on a computer and require no more than a generic computer to generally link the abstract idea to a particular technological environment or field of use (e.g., an online computer network of servers and user devices), with the independent claims amounting to no more than combining the abstract idea with insignificant extra-solution activity including Applicant’s recited operations of “providing” and “performing the action”, as further explained below. As mentioned above, claim elements in addition to the abstract idea arguably include:
using a trained algorithm of a machine learning model (using a trained machine learning model);
providing inputs to the trained algorithm of the machine learning model;
performing the action;
(only Claim 1 and corresponding dependent claims) a system comprising non-transitory memory and a configured processing system coupled to the memory
(only Claim 10 preamble) a computing system; and
(only Claim 19 and corresponding dependent claim) a computer program product comprising a computer-readable storage medium having recorded instructions to enable a processor-based system to perform operations; and each of the recited steps/operations of “providing”.
However, each of these components is recited at a high level of generality that taken individually and in combination perform corresponding generic computer functions of computing ranks, categorizing, ordering, selecting, defining an interval, combining overlaps and approximating targeting, in view of specification paragraphs [0058]–[0064] of Applicant’s disclosure — there is no indication that the combination of elements improves the functioning of a computer or improves any other technology since the additional elements taken individually and collectively merely provide generic computer implementations known to the industry. Furthermore, Examiner notes that none of Applicant’s processes/steps recited in Applicant’s pending claims taken individually and in combination impose a meaningful limit on the claim’s scope since none of recited processes/steps taken individually and in combination involve activity that amounts to more than generic computer functions/activity. Applicant’s steps/processes of computing ranks, categorizing, ordering, selecting, defining an interval, combining overlaps and targeting, as currently recited individually and in combination in Applicant’s claims, are considered to be generic computer functions since they involve generally linking the use of an abstract idea to a particular technological environment or field of use previously known to the industry. Examiner notes that each of the steps of “providing” and “performing” an action/operation encompasses a data output/transmittal function that can be performed by virtually all general purpose computers {see Ultramercial, 772 F.3d at 716‐17; see buySAFE v. Google, 765 F.3d 1350, 1355 (Fed. Cir. 2014); and see Cyberfone Systems v. CNN Interactive Group, 558 Fed. Appx. 988, 993 (Fed. Cir. 2014)}. Also see the “July 2015 Update: Subject Matter Eligibility” document, at page 7, second and sixth bullet points (July 30, 2015) regarding various well‐understood, routine, and conventional functions of a computer. Employing well-known computer functions, such as “using any arbitrary machine learning model” per Applicant’s as-filed specification at paragraph [0056] as well as providing and performing, individually and in combination to execute an abstract idea, even when limiting the use of the idea to one particular environment, does not add significantly more, similar to how limiting the computer-implemented abstract idea in Flook (Parker v. Flook, 437 U.S. 584, 19 U.S.P.Q. 193 (1978)) to petrochemical and oil-refining industries was insufficient. For the reasons discussed above, Applicant’s pending claims do not satisfy Step 2B enunciated in Alice Corp. v. CLS Bank International, 573 U.S. __, 134 S. Ct. 2347 (2014).
Consequently, based upon consideration of all of the relevant factors with respect to each claim as a whole, Claims 1-6, 8-15 and 17-22 are rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter. For information regarding 35 U.S.C. 101, please see Subject Matter Eligibility (SME) guidance and instructional materials at https://www.uspto.gov/patents/laws/examination-policy/subject-matter-eligibility, which includes guidance, memoranda, and updates regarding SME under 35 U.S.C. 101.
Response to Arguments
Applicant’s arguments in the Amendment filed on March 9, 2026, have been fully considered and are not persuasive with respect to 35 U.S.C. § 101.
Applicant's Arguments in the Amendment
(Pages 12-19) Applicant asserts that the pending claims, in view of amendments to the independent claims, are drawn to eligible subject matter under 35 U.S.C. § 101.
Examiner’s Response to Applicant's Arguments
Please see updated/modified § 101 rejections above regarding Applicant’s pending claims being drawn to ineligible subject matter under § 101.
§ 101: Regarding machine learning concepts recited in Applicant’s pending claims, Examiner directs Applicant’s attention to a recent Federal Circuit decision: Recentive Analytics, Inc. v. Fox Corp., Case No. 2023-2437 (Fed. Cir. Apr. 18, 2025), which was decided on April 18, 2025, by the US Court of Appeals for the Federal Circuit (CAFC). A copy of this decision was provided along with the final Office Action dated June 24, 2025. § 101 rejections above make reference to this CAFC case and Examiner encourages Applicant to distinguish its claimed machine learning aspects from this CAFC case.
July 2024: It may be worth noting that Applicant generally argued on pages 13-15 of Applicant’s amendment filed in July 2024, that Applicant’s Claims 1-20 “preserve privacy of users” (page 13 of the July 2024 amendment), “would improve the technical field of preserving privacy of users” (page 14 of the July 2024 amendment), “for example, by identifying users for targeting without a need to leverage their personal identifiers” (bottom of page 14 of the July 2024 amendment with bolding emphases added by Examiner), and “preserving privacy of users” (page 15 of the July 2024 amendment); however, none of Applicant’s pending claims recited preserving privacy of a user or how any element or operation recited in the Applicant’s claims actually accomplishes any of the alleged possible improvements to privacy. Therefore, Applicant's arguments regarding “preserving privacy of users” amounted to no more than a speculative/conclusory allegation that the pending claims improve privacy of users to support an invention under § 101. Applicant's alleged privacy improvement was no more than a desired or intended result that was not required to be performed or achieved based on the claimed language recited. For example, nothing recited in Applicant’s pending claims excluded a scenario where each of Applicant’s recited “relationships” would be unique to a specific individual user (i.e., for classifying or categorizing a specific individual user) and, therefore, each relationship (i.e., between “the values of the features” and “designated operation”) could essentially function as a “personal identifier” when “identifying users for targeting”. Similarly, utilizing “the relationships” for “assigning ranks to respective entities” did not necessarily improve user privacy since a ranking of an entity/user may be used to uniquely identify an individual user/entity and, therefore, each individual ranking could essentially function as a “personal identifier” when “identifying users for targeting”. Examiner noted that use of “relationships” and “ranks”, as was recited in the independent claims, appeared to encompass a situation that actually decreases user privacy over time, especially if the machine learning model learned more (over time) about each user/entity as inputs are received (over time) and used by the machine learning model (over time).
December 2024: In the amendment filed in December 2024, Applicant generally asserted that the pending claims are “preserving privacy of users” (page 15 of December 2024 amendment), and “would improve the technical field of preserving privacy of users” (page 15 of the December 2024 amendment), “for example, by identifying users for targeting without a need to leverage their personal identifiers” (page 15 of the December 2024 amendment with bolding emphases added by Examiner); however, none of Applicant’s December 2024 claims recited preserving privacy of a user or how any element or operation recited in the Applicant’s claims actually accomplished any of the alleged possible improvements to privacy preservation when “approximating personalized targeting”. Therefore, Applicant's December 2024 arguments regarding “preserving privacy of users” amounted to no more than a speculative/conclusory allegation that the December 2024 claims preserved privacy of users to support an invention under § 101. Nonetheless, Examiner acknowledged the field of privacy-preserving data mining (PPDM) is well known. Figure 5 of Applicant’s drawings illustrate how twenty-one (21) entities/users are classified and ranked into “bins” of Bin1, Bin2, Bin3 and Bin4 — which encompasses “personalized targeting” via segmenting a market into groups to be targeted (e.g., target market segmenting), such as for targeted advertising or contextual advertising. Applicant's alleged privacy preservation argument was no more than a desired or intended result that was not required to be performed or achieved based on the language recited in each of Applicant’s December 2024 claims. For example, nothing recited in Applicant’s December 2024 claims excluded a scenario where each of Applicant’s recited “relationships” are unique to a specific individual user (i.e., for classifying or categorizing a specific individual user) and, therefore, each relationship (i.e., between “the values of the features” and “designated operation”) can essentially function as a “personal identifier” when “identifying users for targeting” even “without targeting the targeted entities using an identifier that uniquely identifies an individual targeted entity”, as recited in Applicant’s December 2024 independent claims. Similarly, utilizing “the relationships” for “assigning ranks to respective entities” does not necessarily improve user privacy since a ranking of an entity/user may be used to identify an individual user/entity and, therefore, each individual ranking can essentially function as a “personal identifier” when “identifying users for targeting” even if the ranking does not uniquely identify the individual user/entity. Examiner noted that use of “relationships” and “ranks”, as recited in the December 2024 independent claims, appeared to encompass a situation that actually decreased user privacy over time, especially if the machine learning model “learns” more and more about each user/entity as inputs are received and used by the machine learning model over time. Finally, the alleged improvement in Applicant's December 2024 amendment regarding “privacy of users” amounted to no more than a speculative/conclusory allegation that is not required to be performed or achieved based on the claim language currently recited in Applicant’s December 2024 claims. None of Applicant’s December 2024 claims were required to improve any computer related technology or to improve upon the functioning of the computer itself. Consequently, Applicant's December 2024 arguments constituted no more than a general allegation that the December 2024 claims provides a plurality of improvements to support an invention under § 101.
June 2025: In Applicant’s amendment filed in June 2025, Applicant argued “increases privacy”; however, Examiner noted that each June 2025 independent claim recited “perform the action with regard to the targeted entitles” (underlining emphasis showing a BRI) and that “the targeted entities in compliance with the targeting rule, which increases privacy” encompassed an interpretation where the “the targeted entities in compliance with the targeting rule” is what allegedly “increases privacy” — not the recited “perform the action”, as recited in the June 2025 independent claims. Finally, similar to Recentive Analytics, Inc. v. Fox Corp., Case No. 2023-2437 (Fed. Cir. Apr. 18, 2025), it appeared that Applicant’s June 2025 independent claims “do no more than claim the application of generic machine learning to new data environments, without disclosing improvements to the machine learning model[] to be applied” and, therefore, Applicant’s machine learning concepts, as recited in the June 2025 claims, did not appear to save Applicant’s pending claims from being directed to ineligible subject matter under 35 U.S.C. § 101.
September 2025: In Applicant’s amendment filed in September 2025, Applicant amended each of the independent Claims 1, 10 and 19 to affirmatively recite “increas[ing] privacy of the targeted entities based on the approximat[ing] of the personalized targeting…”, which Applicant argues “increases privacy”. However, it is what Applicant’s claimed subject matter does not do that is associated with Applicant’s alleged improvement. Applicant’s alleged improvement does not result from functions/operations that are positively recited as being performed in each of independent Claims 1, 10 and 19. More specifically, what Applicant’s claimed subject matter does not do —— i.e., Applicant’s claimed subject matter does not “target[] the targeted entities using an identifier that uniquely identifies an individual targeted entity of the targeted entities” per independent claims and/or “without a need to leverage their personal identifiers” (page 15 of September 2025 amendment) —— is what amounts to an absence of a feature (i.e., “targeting… using an identifier that uniquely identifies an individual targeted entity…”). Therefore, it appears that Applicant’s claimed subject matter is relying on an absence of a claimed feature to implicitly/inherently result in Applicant’s alleged improvement to privacy or security (e.g., “privacy of users and/or increasing security of the users” as argued on the 2nd line of page 16 of Applicant’s September 2025 amendment). In contract, what is recited as actually being performed in Applicant’s pending claims appears to have the opposite effect on “user privacy” because Applicant’s recited use of “relationships” and “ranks [that] correspond to respective likelihoods…to perform the designated operation”, as recited in the September 2025 independent claims, appears to encompass a situation that actually decreases user privacy over time — even though abstaining from “targeting… using an identifier that uniquely identifies an individual targeted entity…” would conceivably preserve/maintain privacy in the context of privacy-preserving data mining (PPDM). Compared to doing nothing, such as in an effort to preserve/maintain user privacy, Applicant’s recited use of “relationships” and “ranks [that] correspond to respective likelihoods…to perform the designated operation” appears to actually decrease user privacy over time, especially if Applicant’s recited “machine learning model” learns more and more (over time) about each user/entity as inputs are received (over time) and used by the machine learning model (over time). Consequently, it appears that Applicant’s subject matter is relying on an absence of a claimed feature to allegedly improve “privacy” while actual application/operation of Applicant’s abstract idea in the pending claims (e.g., segmenting a market into groups to be targeted, target market segmenting, targeted advertising, contextual advertising, recited “approximating personalized targeting”, etc.), does not improve any “privacy”. In the context of privacy-preserving data mining (PPDM), Examiner submits that is it well-understood, routine and/or conventional that information that is never received or used necessarily results in privacy with respect to that information (because the information is never available to be used or received). In short, Applicant’s alleged improvement does not result from what operations are actually performed as part of Applicant’s claims subject matter, but rather what Applicant’s claimed subject matter does not do or perform {i.e., Applicant’s claimed subject matter does not “target[]… using an identifier that uniquely identifies an individual targeted entity…” and, similarly, Applicant’s claimed subject matter is “without a need to leverage their personal identifiers” (page 15 of September 2025 amendment)}.
March 2026: In Applicant’s Amendment filed in March 2026, independent Claims 1, 10 and 19 recite either “reduce an amount of resources that is consumed by the…system to target the [the] targeted entities…” and “increasing privacy of a given entity by approximating personalized targeting…”, which Applicant argues “increases privacy”. However, it is what Applicant’s claimed subject matter does not do that is associated with Applicant’s alleged improvements. Applicant’s alleged improvements do not result from functions/operations that are positively recited as being performed in each of independent Claims 1, 10 and 19. More specifically, what Applicant’s claimed subject matter does not do —— i.e., Applicant’s claimed subject matter does not target the targeted entities using “an identifier that uniquely identifies an individual targeted entity of the targeted entities” per independent claims and/or “without a need to leverage their personal identifiers” per page 15 of September 2025 amendment —— is what amounts to an absence of a feature (i.e., absence of targeting using “an identifier that uniquely identifies an individual targeted entity…”). Therefore, it appears that Applicant’s claimed subject matter is relying on an absence of a claimed feature to implicitly/inherently result in Applicant’s alleged improvement of either reducing the system’s consumption of resources, increasing privacy or security (e.g., “privacy of users and/or increasing security of the users” as argued on the 2nd line of page 16 of Applicant’s September 2025 amendment). In contract, what is recited as actually being performed in Applicant’s pending claims appears to have the opposite effect on “user privacy” because Applicant’s recited use of “ranks [that] correspond to respective likelihoods…to perform the designated operation”, as recited in the March 2026 independent claims, appears to encompass a situation that not only increases consumption of resources, but also decreases user privacy over time — even though abstaining from targeting using “an identifier that uniquely identifies an individual targeted entity…” would conceivably preserve/maintain privacy in the context of privacy-preserving data mining (PPDM). Compared to doing nothing, such as in an effort to preserve/maintain user privacy and/or to not consume resources by the system, Applicant’s recited use of “ranks [that] correspond to respective likelihoods…to perform the designated operation” appears to actually decrease user privacy over time (and to actually consume more resources), especially if Applicant’s recited “machine learning model” learns more and more (over time) about each user/entity as inputs are received (over time) and used by the machine learning model (over time). Consequently, it appears that Applicant’s subject matter is relying on an absence of a claimed feature to allegedly improve “privacy” and to “reduce an amount of resources that is consumed” while actual application/operation of Applicant’s abstract idea in the pending claims (e.g., segmenting a market into groups to be targeted, target market segmenting, targeted advertising, contextual advertising, recited “approximating personalized targeting”, etc.), does not improve any “privacy” or “resource consumption”. In the context of privacy-preserving data mining (PPDM), Examiner submits that is it well-understood, routine and/or conventional that information that is never received or used necessarily results in privacy with respect to that information (because the information is never available to be used or received). In short, Applicant’s alleged improvements do not result from what operations are actually performed as part of Applicant’s claims subject matter, but rather what Applicant’s claimed subject matter does not do or perform {i.e., Applicant’s claimed subject matter does not target using “an identifier that uniquely identifies an individual targeted entity…” and, similarly, Applicant’s claimed subject matter is “without a need to leverage their personal identifiers” (page 15 of September 2025 amendment)}.
In addition, similar to Recentive Analytics, Inc. v. Fox Corp., Case No. 2023-2437 (Fed. Cir. Apr. 18, 2025), it appears that Applicant’s most recent independent claims “do no more than claim the application of generic machine learning to new data environments, without disclosing improvements to the machine learning model[] to be applied” and, therefore, Applicant’s machine learning concepts, as currently recited, do not appear to save Applicant’s pending claims from being directed to ineligible subject matter under 35 U.S.C. § 101.
In view of Applicant’s arguments and above § 101 analysis, Examiner is unpersuaded and, therefore, maintains/asserts rejections to Claims 1-6, 8-15 and 17-22 under 35 U.S.C. § 101.
Conclusion
The following references are considered pertinent to Applicant's disclosure, and are being made of record albeit the references are not relied upon as a basis for a rejection in this Office action:
U.S. Patent Application Publication No. 2024/0054391 of Guha Thakurta et al. (hereinafter “Guha Thakurta”) for “privacy-enhanced training and deployment of machine learning models that operate on data resident on client-side device(s) and server-side device(s)” —Guha Thakurta at ¶ [0001]; and “…content personalization which can generally involve training and deploying a machine learning model that predicts a particular digital component to provide for display on a client device, where the model uses both client-side data and server-side data, such as attributes of available digital components, which are stored at a server, and data about user's interests and preferences, which are stored at a client device” —Guha Thakurta at ¶ [0003].
U.S. Patent Application Publication No. 2022/0391778 of Green et al. (hereinafter “Green”) for “privacy-preserving federated learning techniques which enables the use of locally stored data to learn entity embeddings” —Green at ¶ [0001].
U.S. Patent Application Publication No. 2023/0274183 of Mauser et al. (hereinafter “Mauser”) for “PROCESSING OF MACHINE LEARNING MODELING DATA TO IMPROVE ACCURACY OF CATEGORIZATION” —Title of Mauser.
U.S. Patent Application Publication No. 2023/0205915 of Wang et al. (hereinafter “Wang”) for “PRIVACY PRESERVING MACHINE LEARNING FOR CONTENT DISTRIBUTION AND ANALYSIS” —Title of Wang.
U.S. Patent Application Publication No. 2022/0171874 of Leif-Nissen LUNDBÆK (hereinafter “LUNDBÆK ‘874”).
U.S. Patent Application Publication No. 2022/0171873 of Leif-Nissen LUNDBÆK (hereinafter “LUNDBÆK ‘873”).
U.S. Patent Application Publication No. 2021/0256309 of HUTH et al. (hereinafter “Huth”).
U.S. Patent Application Publication No. 2021/0224862 of Taifi et al. (hereinafter “Taifi”) for “generating a machine learning model to rank the targetable users and building a new segment based on users ranked by the model” —Abstract of Taifi; “grouping online users in audience segments. These segments are used by marketers, ad traders, and campaign managers to deploy personalized ad targeting” —Taifi at ¶ [0150]; “the top N ranked targetable users are selected and combined into the new audience segment, where N is any suitable number. The system then attaches the new segment to the line item and begins targeting users of the new segment, along with users of previously attached segments of the line item. In this manner, more users are available to receive impressions of the line item and the delivery of impressions can be better controlled to match the delivery goals of the campaign and line item” —Taifi at ¶ [0157]; and “Given a set of advertisement campaigns C={c1, . . . , c|C|}, and a set of segments S={s1, . . . , s|s|}, the system and method will retrieve segments that are most similar to other sets of segments. This requires a similarity metric between segments. The system and method extract this similarity metric by embedding the segments in a learned dense low dimensional space using the relationship between the campaigns and the segments” —Taifi at ¶ [0161]; and “dataset was fully anonymized to remove any member level data” —Taifi at ¶ [0212].
U.S. Patent Application Publication No. 2021/0216923 of RAMA et al. (hereinafter “Rama ‘923”) for “select the top N ranked supply agents 218 for each demand agent, where the rankings of the supply agents are based on the predicted probability of future transactions with the supply agents for each demand agent. The variable N for the supply agent selector 206 may be user configurable. As an example, N may be configured as the top 2% of the supply agents. However, this can always be tuned as a hyper-parameter to be optimized for the model. In an embodiment, the supply agent selector 206 sorts the predicted probabilities of supply agents for each demand agent in descending order and then selects the first N supply agents from the top of the list” —Rama ‘923 at ¶ [0033].
U.S. Patent Application Publication No. 2021/0073672 of Shi et al. (“Shi”).
U.S. Patent Application Publication No. 2019/0102681 of Roberts et al. (hereinafter “Roberts”) for “receive partially anonymized or non-anonymized data from one or more data providers and can obscure or eliminate fields in individual records according to data-privacy rules and/or can aggregate field values across a user sub-population to comply with data-privacy rules” —Roberts at ¶ [0028].
U.S. Patent Application Publication No. 2017/0330231 of WAYNE et al. (“Wayne”).
U.S. Patent Application Publication No. 2015/0186938 of Zhang et al. (hereinafter “Zhang”) for “feature values are sorted into pre-defined bins, and in another embodiment clustering or other means are used to identify bins for clustering (or otherwise grouping) features, such as based on quantized values of the features. For example, in an embodiment the TF of a term are binned, such as for example by quantizing the TF feature as 0 to 100. In this manner, each bin becomes a feature usable for the machine learning scorer. Thus in an embodiment, tokens (terms) are grouped together. In another embodiment, the IDF feature is binned” —Zhang at ¶ [0054].
U.S. Patent Application Publication No. 2013/0124298 of Li et al. (hereinafter “Li”) for “analyzes other engagement information about the users in the training cluster, such as expressing interest in a page on the social networking system (becoming a "fan" of a page or "liking" a page) and installing an application on the social networking system. Users may be sorted by their distribution of engagement with advertisers on the social networking system into bins in a similar way as described above. For example, users may be segmented by the number of pages on the social networking system that the users have expressed interest in or by the number of applications the users have installed on the social networking system. Once the user model has been generated for the advertiser based on the training cluster of users and applied to the total population of users, the same percentage of users may be selected from the bins to normalize the engagement information of the users in the training cluster” —Li at ¶ [0034]; and “A machine learning module 310 is used in the user model generating module 108 to select features for user models generated for advertisers. In one embodiment, a social networking system 100 uses a machine learning algorithm to analyze user characteristics of a training cluster of users that have engaged with an advertiser by clicking on an advertisement, for example. The machine learning module 310 may select user characteristics as features for the user model for the advertiser, such as keyword features, demographic features, and advertiser click features, using at least one machine learning algorithm. In another embodiment, a machine learning algorithm may be used to optimize the selected features for a user model based on conversion rates of advertisements targeted to users identified from the user model. A selected feature in a user model may be removed based on a lack of engagement by users targeted by the advertiser based on the user model that exhibit the selected feature. For example, a selected feature for a user model may include a high affinity score for Starbucks Coffee. However, if users exhibiting a high affinity score for Starbucks Coffee do not engage with the advertisement in expected numbers, then the machine learning algorithm may deselect the feature in the user model” —Li at ¶ [0038].
U.S. Patent Application Publication No. 2011/0264513 of Adwait Ratnaparkhi (hereinafter “Ratnaparkhi”) for “selecting a plurality of features from the training model dataset and calculating a click probability for a subject advertisement to be clicked by a user from a page, the calculating operations using features of the page that is to be presented to the user. Embodiments include mapping a particular query to one of the targeting categories and then presenting the subject advertisement selected on the basis of the value of the click probability. Normalization scales down the value of the click probabilities to filter out false positive categories” —Abstract of Ratnaparkhi.
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Mathew Syrowik whose telephone number is 313-446-4862. The examiner can normally be reached on Monday through Friday 8:30 AM to 4:00 PM (Eastern Time). If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Waseem Ashraf, can be reached at telephone number 517-270-3948. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from Patent Center. Status information of published applications may be obtained from Patent Center. Status information of unpublished applications is available through Patent Center for authorized users only. Should you have questions about access to Patent Center, please contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free) or by email at EBC@uspto.gov. Examiner interviews are available via telephone or video conference using a USPTO supplied web-based collaboration tool. To schedule an interview, please email Mathew.Syrowik@USPTO.gov or applicant may use the USPTO Automated Interview Request (AIR) Form at https://www.uspto.gov/patents/uspto-automated-interview-request-air-form. For additional information or questions, please contact the Inventors Assistance Center at 1-800-786-9199 (toll free), 571-272-1000 (local), or 1-800-877-8339 (TDD/TTY).
/Mathew Syrowik/Primary Examiner, Art Unit 3621
Prosecution Insights