DETAILED ACTION
Response to Amendment
This action is in response to the response to the amendment filed on 06/30/2026. Claims 1, 11, and 12 have been amended and claims 13-16 have been newly added. Claims 1-16 are pending and currently under consideration for patentability.
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 .
Inventorship
This application currently names joint inventors. In considering patentability of the claims under pre-AIA 35 U.S.C. 103(a), the examiner presumes that the subject matter of the various claims was commonly owned at the time any inventions covered therein were made absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and invention dates of each claim that was not commonly owned at the time a later invention was made in order for the examiner to consider the applicability of pre-AIA 35 U.S.C. 103(c) and potential pre-AIA 35 U.S.C. 102(e), (f) or (g) prior art under pre-AIA 35 U.S.C. 103(a).
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 1-16 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. The claims are directed to a judicial exception (i.e., a law of nature, natural phenomenon, or abstract idea) without significantly more.
Step 1: In a test for patent subject matter eligibility, claims 1-16 are found to be in accordance with Step 1 (see 2019 Revised Patent Subject Matter Eligibility), as they are related to a process, machine, manufacture, or composition of matter. Claims 1-10, 13-16 recite a system, claim 11 recites a method, and claim 12 recites a non-transitory computer-readable medium. When assessed under Step 2A, Prong I, they are found to be directed towards an abstract idea. The rationale for this finding is explained below:
Step 2A, Prong I: Under Step 2A, Prong I, claims 1, 11, and 12 are directed to an abstract idea without significantly more, as they all recite a judicial exception. Claims 1, 11, and 12 recite limitations directed to the abstract idea including “acquiring target information including information of a specific target and target; estimating the impression that the user is estimated to have on the specific target from among a plurality of impressions included in an impression group; determining the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group; generating merge information by merging a result of the estimating and a result of the determining; and generating a prompt on a basis of the merge information.” These further limitations are not seen as any more than the judicial exception. Claims 1, 11, and 12 recite additional limitations including “using generative artificial intelligence on a basis of the target information and/or user action information indicating an action of the user regarding the specific target; and inputting the prompt to the generative artificial intelligence to cause the generative artificial intelligence to output improvement information including information indicating improvement content related to the target information.” The claims are considered to be an abstract idea under certain methods of organizing human activity because the claims are directed to commercial or legal interactions (including agreements in the form of contracts; legal obligations; advertising, marketing or sales activities or behaviors; business relations) and managing personal behavior or relationships or interactions between people (including social activities, teaching, and following rules or instructions) such as improving content based on estimated and determined information. The claims are also considered to be an abstract idea under Mental Processes such as concepts performed in the human mind (including an observation, evaluation, judgment, opinion) because the claims are directed to acquiring information (i.e. target information, target impression information, and user impression information), estimating information (i.e. estimating user impression), determining information (i.e. if the impression occurred), and generating information (i.e. merge information). Therefore, under Step 2A, Prong I, claims 1, 11, and 12 are directed towards an abstract idea.
Step 2A, Prong II: Step 2A, Prong II is to determine whether any claim recites any additional element that integrate the judicial exception (abstract idea) into a practical application. Claims 1, 11, and 12 recite additional limitations including “using generative artificial intelligence on a basis of the target information and/or user action information indicating an action of the user regarding the specific target; and inputting the prompt to the generative artificial intelligence to cause the generative artificial intelligence to output improvement information including information indicating improvement content related to the target information.” “Using generative artificial intelligence” appears twice, both times as a black-box tool invoked to perform the abstract mental step (estimate an impression; determine an impression) — no technical detail is given about how the generative AI accomplishes this (no model architecture, no training method, no specific data representation). “Generating a prompt on a basis of the merge information” and “inputting the prompt to the generative artificial intelligence” is simply describing the generic, well-known mechanics of using an LLM — assemble text, send it to the model, get text back. This is precisely the kind of generic, off-the-shelf, functionally-claimed use of AI that recent case law has found insufficient to integrate an abstract idea into a practical application. These additional limitations are seen as adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f). Accordingly, alone, and in combination, these additional elements are seen as using a computer or tool to perform an abstract idea, adding insignificant-extra-solution activity to the judicial exception. They do no more than link the judicial exception to a particular technological environment or field of use (i.e. generative AI) and therefore do not integrate the abstract idea into a practical application. Recentive Analytics, Inc. v. Fox Corp., 134 F.4th 1205 (Fed. Cir. 2025) — the Federal Circuit held claims reciting the use of machine learning to generate network maps and event schedules were ineligible, because the claims merely applied generic, unspecified machine learning techniques to a new problem/data domain, without claiming any improvement to how the machine learning model itself works. The court emphasized that “training a machine learning model” and “applying” it to generate an output, without more, does not transform an abstract idea into a patent-eligible invention — result-oriented, functional claiming of AI is not itself an inventive technical solution. This claim has the same defect: it claims the result (“cause the generative artificial intelligence to output improvement information”) without claiming how the generative AI is technically adapted, improved, or configured to achieve that result better than any generic LLM would if simply prompted with the same merged data. There’s no claimed improvement to: the underlying AI model architecture, the computer’s functioning, a technical field, or any particular machine or transformation of matter. The courts decided that although the additional elements did limit the use of the abstract idea, the court explained that this type of limitation merely confines the use of the abstract idea to a particular technological environment and this fails to add an inventive concept to the claims (See Affinity Labs of Texas v. DirecTV, LLC,). Under Step 2A, Prong II, these claims remain directed towards an abstract idea.
Step 2B: Claims 1, 11, and 12 recite additional limitations including “using generative artificial intelligence on a basis of the target information and/or user action information indicating an action of the user regarding the specific target; and inputting the prompt to the generative artificial intelligence to cause the generative artificial intelligence to output improvement information including information indicating improvement content related to the target information.” These additional limitations do not integrate the judicial exception (abstract idea) into a practical application because of the analysis provided in Step 2A, Prong II. Claims 1, 11, and 12 do not include additional elements or a combination of elements that result in the claims amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements listed amount to no more than mere instructions to apply an exception using a generic computer component. In addition, the applicant’s specifications describe a “general purpose graphic processing unit”, pages 44-45, for implementing the computer/apparatus, which do not amount to significantly more than the abstract idea of itself, which is not enough to transform an abstract idea into eligible subject matter. Furthermore, there is no improvement in the functioning of the computer or technological field, and there is no transformation of subject matter into a different state. Under Step 2B in a test for patent subject matter eligibility, these claims are not patent eligible.
Dependent claims 2-10 and 13-16 further recite the system of claim 1. Dependent claims 2-10 and 13-16 when analyzed as a whole are held to be patent ineligible under 35 U.S.C. 101 because the additional recited limitation fail to establish that the claims are not directed to an abstract idea:
Under Step 2A, Prong I, these additional claims only further narrow the abstract idea set forth in claims 1, 11, and 12. For example, claims 2-10 and 13-16 describe the limitations for improving content based on target information, target impression information, and user impression information – which is only further narrowing the scope of the abstract idea recited in the independent claims.
Under Step 2A, Prong II, for dependent claims 2-10 and 13-16, there are no additional elements introduced. For example, dependent claims 4 recites additional limitations including “using generative AI” and dependent claim 10 recites additional limitations including “to generative AI” and “from the generative AI”. These additional limitations are seen as adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f). Accordingly, alone, and in combination, these additional elements are seen as using a computer or tool to perform an abstract idea, adding insignificant-extra-solution activity to the judicial exception. They do no more than link the judicial exception to a particular technological environment or field of use (i.e. generative AI) and therefore do not integrate the abstract idea into a practical application. Thus, they do not present integration into a practical application, or amount to significantly more.
Under Step 2B, the dependent claims do not include any additional elements that are sufficient to amount to significantly more than the judicial exception. Additionally, there is no improvement in the functioning of the computer or technological field, and there is no transformation of subject matter into a different state. As discussed above with respect to integration of the abstract idea into a practical application, the additional claims do not provide any additional elements that would amount to significantly more than the judicial exception. Under Step 2B, these claims are not patent eligible.
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.
Claim(s) 1-5 and 7-12 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Publication 2020/0034874 to Narayan in view of U.S. Publication 2025/0068893 to Bradea.
With respect to Claim 1:
Narayan teaches:
An information processing apparatus comprising: at least one memory; and at least one processor coupled to the at least one memory and configured to execute (Narayan: ¶ [0117]):
a target related information acquisition unit that acquires target information including information of a specific target (i.e. receiving target information including type of product/brand being advertised or target demographic) (Narayan: ¶¶ [0031] [0032] “For example, the computer system can store a predefined "viewability" model configured to intake a series of historical session containers of a user and to output a probability that the user will scroll down to an advertisement inserted into a webpage and that a minimum proportion of this advertisement will be rendered on the user's computing device for at least a minimum duration of time based on these engagement data. The viewability model can also: intake metadata of an advertisement, such as the format of the advertisement ( e.g., static or interactive with video, catalog, virtual reality, or hotspot content) and a type of brand or product advertised; and output a probability that a user will scroll down to this advertisement inserted into a webpage viewed on the user's computing device and that the minimum proportion of this advertisement will be rendered on the user's computing device for at least the minimum duration of time based on historical user engagement data and these advertisement metadata…The computer system can similarly implement other intent models, such as: a conversion model that outputs a probability that a user will convert through an advertisement served to a webpage accessed on the user's computing device; a click-through model that outputs a probability that a user will click on an advertisement; a scroll interaction model that outputs a probability that a user will scroll back and forth over an advertisement at least a minimum number of times; a hotspot model that outputs a probability that a user will select at least a minimum number of hotspots within an interactive advertisement; a swipe model that outputs a probability that a user will swipe laterally through content within an advertisement; a virtual reality model that outputs a probability that a user will manipulate a virtual advertisement environment within an advertisement to at least a minimum degree; a video model that outputs a probability that a user will view at least a minimum duration or proportion of a video within an advertisement; and/or a brand lift model that outputs a probability that a user will exhibit at least a threshold increase in brand recognition after an advertisement is served to the user's computing device; etc.” Furthermore, as cited in ¶ [0038] “In one implementation, the computer system can aggregate a population of users who may be candidates for serving an advertisement in the new campaign, such as by user demographic (e.g., age, gender), location, and/or other characteristics specified by the new advertising campaign.”);
an estimation unit that estimates the impression that the user is estimated to have on the specific target from among a plurality of impressions included in an impression group [[using generative artificial intelligence on a basis of the target information]] (i.e. intent model predicts probability of user interaction or engagement, wherein classifications include impression groups such as highly-engaged) (Narayan: ¶¶ [0032] [0033] “The computer system can similarly implement other intent models, such as: a conversion model that outputs a probability that a user will convert through an advertisement served to a webpage accessed on the user's computing device; a click-through model that outputs a probability that a user will click on an advertisement; a scroll interaction model that outputs a probability that a user will scroll back and forth over an advertisement at least a minimum number of times; a hotspot model that outputs a probability that a user will select at least a minimum number of hotspots within an interactive advertisement; a swipe model that outputs a probability that a user will swipe laterally through content within an advertisement; a virtual reality model that outputs a probability that a user will manipulate a virtual advertisement environment within an advertisement to at least a minimum degree; a video model that outputs a probability that a user will view at least a minimum duration or proportion of a video within an advertisement; and/or a brand lift model that outputs a probability that a user will exhibit at least a threshold increase in brand recognition after an advertisement is served to the user's computing device; etc…In one example, the computer system implements an intent model that correlates user interactions to likelihood that a user will perform a downstream action separate from the target interactions for the advertisement, such as: make a physical or digital purchase; exhibit greater brand recognition; spend more time within an advertiser's website; or exhibit greater lifetime value as a customer of the advertiser. In this example, the computer system can serve brand lift, product purchase, and/or other surveys to these users over time, link results of these surveys to related advertisements previously served to these users, and then implement linear regression, artificial intelligence, a convolutional neural network, or other analysis techniques to develop an intent model linking advertising content previously served to these users, placement of these advertisements, user characteristics, and user interactions with advertisements to these outcomes indicated in these surveys.” Furthermore, as cited in ¶¶ [0067] [0068] “);
a determination unit that determines the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group [[using the generative artificial intelligence on a basis of user action information indicating an action of the user regarding the specific target]] (i.e. receiving actual engagement metrics user determined to have with advertisement) (Narayan: ¶ [0017] “Visual elements served to the user in this population can include iframe elements loaded with static, video, and or dynamic (e.g., responsive) advertising content that can be configured to regularly record various direct and indirect engagement metrics, such as: the position of the advertisement within a viewing window rendered on a display of a computing device associated with the user; a number of pixels of the advertisement currently in view in the viewing window; clicks over the advertisement; touch events over the advertisement (i.e., inside of the visual element); touch events outside the advertisement (i.e., outside of the visual element) while the advertisement is in view in the viewing window; vertical scroll events that move the advertisement within the viewing window; horizontal swipes over the advertisement; hotspot selections within the advertisement; video plays, pauses, and resumes within the advertisement; and metadata of the webpage containing the advertisement; etc. For example, a visual element inserted into a webpage rendered within a web browser executing on a user's mobile computing device can regularly collect these engagement data and return these engagement data to the computer system.”).
Narayan does not explicitly disclose an estimation unit that estimates the impression that the user is estimated to have on the specific target from among a plurality of impressions included in an impression group using generative artificial intelligence on a basis of the target information; a determination unit that determines the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group using the generative artificial intelligence on a basis of user action information indicating an action of the user regarding the specific target; and a generation unit that: generates merge information by merging a result of the estimation by the estimation unit and a result of the determination by the determination unit, generates a prompt on a basis of the merge information, and inputs the prompt to the generative artificial intelligence to cause the generative artificial intelligence to output improvement information including indicating improvement content related to the target information.
However, Bradea further discloses:
an estimation unit that estimates [[the impression that the user is estimated to have on the specific target from among a plurality of impressions included in an impression group]] using generative artificial intelligence on a basis of the target information (i.e. generative AI is trained to make predictions on data in order to determine outputs based on attributes and segment/event information of the user or estimate/infer attributes 102 using generative AI) (Bradea: ¶ [0028] “As used herein, the term "generative artificial intelligence (AI)" refers to hardware and/or software module(s) including program code that can create new content or make predictions based on the data it has been trained on. For example, a generative AI may include a chatbot that can accept arbitrary textual prompts and then generate novel responses. Generative AI systems can generate text, graphics, video, audio outputs, combinations thereof, or other output formats.” Furthermore, as cited in ¶ [0022] “The personalization module then determines a tuning parameter for a generative AI model, wherein the tuning parameter controls the randomness of the output of the generative AI model. For example, some generative AI models accept a temperature parameter for modulating the randomness of the output. A higher temperature corresponds to more randomness in the output given a prompt, where a lower temperature corresponds to less randomness given the same prompt. The personalization module may use information about the user interactions, including the events associated with those interactions, to determine the tuning parameter. For instance, the personalization module may use counts of events associated with the user or the product to determine the tuning parameter.” Furthermore, as cited in ¶¶ [0037] [0038] “For example, the content generation subsystem 130 may receive attributes 102, including information about the user and/or about the subject of the digital content ( e.g., a product)…The content generation subsystem 130 may also receive information about one or more segments to which the user belongs. For instance, the attributes 102 received by the content generation subsystem 130 may indicate that a web shopper is a 37 year-old female and that the subject of the content is a grey fleece. Identifying information about the user can be used to determine segments to which the user belongs. The example segment 104 describes its members as being Samsung users who enjoy mounting activities such as trekking. Segments may be determined on the basis of previously obtained information, retrieved from third-party sources or other databases, or inferred from user information and behaviors, among other possible mechanisms.”);
a determination unit that determines the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group using the generative artificial intelligence on a basis of user action information indicating an action of the user regarding the specific target (i.e. determines event information such as actions taken by the user using the tuning parameter of the generative AI) (Bradea: ¶¶ [0039] [0040] “The content generation system 130 may also receive information about events 103 that have occurred with respect to the content, such as an indication that a user interacted with content displayed on a web page. Events 103 may include actions taken by users with respect to particular content, with each event 103 indicative of a positive predisposition towards the subject matter of the content. For instance, events like the addition of an item to a virtual shopping cart or wish list may be indicative of an inclination to purchase the item and can thus be used as a signal for the generation of personalized content. In this example, events 103 corresponding to a positive predisposition are used, but any type of event or user behavior may be relevant to the development of personalized content, including neutral or negative events…The content generation subsystem 130, based on some or all of these inputs, can determine a tuning parameter 160 for the generative AI 150. In some examples, the tuning parameter may control the randomness of the output of the generative AI 150. However, other aspects of the generative AI 150 may be controlled with the tuning parameter 160 such as exploration, precision and recall, the speed vs. accuracy tradeoff, the number of neural network layers, and so on. Examples of tuning parameters 160 include presence penalties, frequency penalties, stopping criteria, beam searches, reward models, soft prompts, and others.”); and
a generation unit that: generates merge information by merging a result of the estimation by the estimation unit and a result of the determination by the determination unit (i.e. attributes 102 such as target/content information, corresponding to the estimation-side input is combined with events 103 such as user-action-driven information, corresponding to the determination-side result) (Bradea: ¶ [0042] “The content generation subsystem 130 can generate a prompt based on some or all of these inputs. For example, the content generation subsystem 130 may combine the information from the attributes 102, information about the events 103, and the user segment information 104 to generate a human-readable prompt. The prompt may be in the form of a sentence, paragraph, or paragraphs, and phrased as if it were a request to another human being. In some examples, the prompt may be in a semi-structured format.”),
generates a prompt on a basis of the merge information (i.e. generates the prompt by incorporating both target information such as product description, “grey hooded fleece” and user-action-derived information such as “added this item to her favorites”). (Bradea: ¶ [0047] “For example, a typical example prompt generated based on the product description 170, the attributes 102, information about the events 103, the user segment information 104, and the tuning parameter 160 may be: "Generate a new product description for a grey hooded fleece. The description is currently 'Experience Comfort and Style with our Hooded Fleece-the Ultimate Hoodie for the Active Woman!' The user is a 37 year old female. She is a Samsung phone user and enjoys mountain activities and trekking. She has added this item to her favorites. 3 other people have added this item to their favorites or purchased the item in the last month." This prompt may be sent to the generative AI 150 along with instructions to use a specific temperature. For example, the prompt and temperature may be sent to an API provided by generative AI 150, in which both the prompt and the temperature are included as objects in a data structure, like a JSON object.”), and
inputs the prompt to the generative artificial intelligence to cause the generative artificial intelligence to output improvement information including indicating improvement content related to the target information (i.e. input prompt into generative AI to output new/updated content based on prompts tailored to improve the personalization of content) (Bradea: ¶ [0043] “The content generation subsystem 130 may then input the generated prompt to the generative AI 150. Generative AI 150 includes a class of technologies that can generate original content on the basis of a substantial volume of training data. For example, generative AI 150 technologies exist for generating original text, graphical content, and even audio content. Well-known examples of generative AI 150 technologies for the generation of text include ChatGPT, Claude, Language Model for Dialogue Applications (LaMDA), Bard, and many others. Examples of AI technologies for the generation of graphical content include DALL-E, MidJourney, and Adobe Sensei.” Furthermore, as cited in ¶ [0048] “The content generation subsystem 130 may receive, from the generative AI 150, personalized content responsive to the tuning parameter and the prompt. For example, in response to the previous example prompt, the generative AI 150 may respond, "Discover Elegance and Ease in our Fleece Hoodie-the Supreme Sweatshirt for Hiking." The updated description can be extracted from the response and used to update product description 170 contained in dynamic content field 175 to become updated product description 180, as described next.” Furthermore, as cited in ¶ [0037] “The content generation subsystem 130 may be configured to cause a generative AI 150 to generate new or updated content based on prompts tailored to improve the personalization of content for a particular user or group of users. For example, the content generation subsystem 130 may receive attributes 102, including information about the user and/or about the subject of the digital content ( e.g., a product).”).
Therefore, it would have been obvious to one of ordinary skill in the art, at the time the invention was made, to add Bradea’s estimation unit that estimates the impression that the user is estimated to have on the specific target from among a plurality of impressions included in an impression group using generative artificial intelligence on a basis of the target information; a determination unit that determines the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group using the generative artificial intelligence on a basis of user action information indicating an action of the user regarding the specific target; and a generation unit that: generates merge information by merging a result of the estimation by the estimation unit and a result of the determination by the determination unit, generates a prompt on a basis of the merge information, and inputs the prompt to the generative artificial intelligence to cause the generative artificial intelligence to output improvement information including indicating improvement content related to the target information to Narayan’s target related information acquisition unit that acquires target information including information of a specific target; an estimation unit that estimates the impression that the user is estimated to have on the specific target from among a plurality of impressions included in an impression group; and a determination unit that determines the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group. One of ordinary skill in the art would have been motivated to do so because “Such precision marketing in near-real-time was not possible using existing tools, much less scalable. Thus, the techniques of the present disclosure can result in a higher conversion rate due to the highly personalized experience.” (Bradea: ¶ [0027]).
With respect to Claims 11 and 12:
All limitations as recited have been analyzed and rejected to claim 1. Claim 11 recites “An information processing method executed by a computer, the method comprising:” the steps of system claim 1. Claim 12 recites “A non-transitory computer readable storage medium having stored therein an information processing program for causing a computer to execute:” (Narayan: ¶ [0117]) the steps of system claim 1. Claims 11 and 12 do not teach or define any new limitations beyond claim 1. Therefore they are rejected under the same rationale.
With respect to Claim 2:
Narayan teaches:
The information processing apparatus according to claim 1, comprising: a reception unit that receives improvement directionality information indicating directionality of improvement of the target information, wherein the generation unit generates the improvement information on a basis of information further including the improvement directionality information received by the reception unit (i.e. receive characteristic of level of success or improvement directionality which generates information regarding level of success information of an ideal advertising campaign) (Narayan: ¶¶ [0075] [0076] “For example, the computer system can interpret a wide funnel top, narrow funnel center, and wide funnel end as a "polarizing ad" that yields high engagement when served to an interested party but otherwise yields minimal engagement; the computer system then automatically prompt a campaign manager to modify the advertisement to reduce polarization and thus engage for more users. Alternatively, the computer system can automatically isolate common user and environment characteristics of advertisement sessions proximal the funnel end and selectively target the advertisement to users exhibiting these characteristics in similar environments. In another example, the computer system can interpret a wide funnel top, wide funnel center, and narrow funnel end as a "promising ad" that yields high initial user engagement but fails to push users to a CTA; the computer system then automatically prompt a campaign manager to modify the CTA in the advertisement in order to push more users from a engaged state to a highly-engaged state…In another implementation, the computer system can store a set of funnel visualization templates depicting funnel characteristics of advertising campaigns exhibiting different levels of success, such as: a highly-successful campaign ( or "ideal advertising campaign") with a high ratio of total users to highly-engaged users; a moderately-successful campaign with a moderate ratio of total users to highly-engaged users; a minimally-successful campaign with a low ratio of total users to highly-engaged users; a polarizing campaign with a low ratio of total users to engaged users; a promising campaign with a high ratio of total users to engaged users and a low ratio of engaged users to highly-engaged users. In this implementation, the computer system can identify a funnel visualization template nearest to the funnel visualization generated for an advertising campaign, scale the funnel visualization template to the funnel visualization, overlay this funnel visualization template over the funnel visualization, and present this composite funnel visualization to the campaign manager. Alternatively, the computer system can store a single funnel visualization template (e.g., for an ideal advertising campaign), scale the funnel visualization template to the funnel visualization, overlay this funnel visualization template over the funnel visualization, and present this composite funnel visualization to the campaign manager in order to indicate to the campaign manager how the advertising campaign is tracking relative to an ideal advertising campaign.”).
With respect to Claim 3:
Narayan teaches:
The information processing apparatus according to claim 2, wherein the directionality of improvement of the target information is directionality of improvement to be closer to an impression that the user is estimated to have and directionality of improvement to be closer to an impression that the user is determined to have had (i.e. level of success or improvement directionality is how the ideal campaign or impression user is estimated to have is tracking relative to the actual campaign or impression user is determined to have had) (Narayan: ¶¶ [0075] [0076] “For example, the computer system can interpret a wide funnel top, narrow funnel center, and wide funnel end as a "polarizing ad" that yields high engagement when served to an interested party but otherwise yields minimal engagement; the computer system then automatically prompt a campaign manager to modify the advertisement to reduce polarization and thus engage for more users. Alternatively, the computer system can automatically isolate common user and environment characteristics of advertisement sessions proximal the funnel end and selectively target the advertisement to users exhibiting these characteristics in similar environments. In another example, the computer system can interpret a wide funnel top, wide funnel center, and narrow funnel end as a "promising ad" that yields high initial user engagement but fails to push users to a CTA; the computer system then automatically prompt a campaign manager to modify the CTA in the advertisement in order to push more users from a engaged state to a highly-engaged state…In another implementation, the computer system can store a set of funnel visualization templates depicting funnel characteristics of advertising campaigns exhibiting different levels of success, such as: a highly-successful campaign ( or "ideal advertising campaign") with a high ratio of total users to highly-engaged users; a moderately-successful campaign with a moderate ratio of total users to highly-engaged users; a minimally-successful campaign with a low ratio of total users to highly-engaged users; a polarizing campaign with a low ratio of total users to engaged users; a promising campaign with a high ratio of total users to engaged users and a low ratio of engaged users to highly-engaged users. In this implementation, the computer system can identify a funnel visualization template nearest to the funnel visualization generated for an advertising campaign, scale the funnel visualization template to the funnel visualization, overlay this funnel visualization template over the funnel visualization, and present this composite funnel visualization to the campaign manager. Alternatively, the computer system can store a single funnel visualization template (e.g., for an ideal advertising campaign), scale the funnel visualization template to the funnel visualization, overlay this funnel visualization template over the funnel visualization, and present this composite funnel visualization to the campaign manager in order to indicate to the campaign manager how the advertising campaign is tracking relative to an ideal advertising campaign.”).
With respect to Claim 4:
Narayan teaches:
The information processing apparatus according to claim 1, wherein the generation unit generates the improvement information using generative Al (i.e. generate advertisement recommendations/improvement/performance information using artificial intelligence) (Narayan: ¶ [0113] “The remote computer system (or other computer system) can then implement linear regression, artificial intelligence, a convolutional neural network, or other analysis techniques to derive correlations between: engagement layer characteristics, mobile advertisement characteristics, user characteristics, and/or environment characteristics; and outcomes of composite mobile advertisements constructed from mobile/engagement layer pairs. The remote computer system can similarly derive correlations between these characteristics and outcomes of mobile advertisements served to users without engagement layers. For example, the remote computer system can identify: mobile advertisement format and engagement layer animation combinations that correlate with higher frequency instances of scroll events over an advertisement; engagement layers that correlate with higher frequency of conversions when placed in advertisements at the bottom of a webpage; and/or CTA placement and animations in an engagement layer that correlate with higher frequency of brand lift when paired with mobile advertisements advertising a particular category of product ( e.g., menswear, vehicles).”).
With respect to Claim 5:
Narayan teaches:
The information processing apparatus according to claim 1, comprising: a determination unit that determines an impression that the user has had on a basis of information of the user regarding the specific target (i.e. engagement includes user’s interaction with specific product in advertisement) (Narayan: ¶ [0094] “The engagement layer can also include a call to action (hereinafter "CTA"), such as a textual statement or icon configured to persuade a user to perform a particular task, such as purchasing a product, signing up for a news letter, or clicking-through to a landing page for a brand or product. For example, the engagement layer can include a generic CTA ( e.g., "Click to learn more>>>") with an empty link, and an advertisement receiving this engagement layer can tie the CTA in the engagement layer to a link-to an external webpage-contained in the mobile advertisement.”).
With respect to Claim 7:
Narayan teaches:
The information processing apparatus according to claim 1, comprising: an estimation unit that estimates an impression that the user is estimated to have on a basis of the target information (i.e. determining probability of different impressions estimated to occur based on advertisement) (Narayan: ¶ [0032] “The computer system can similarly implement other intent models, such as: a conversion model that outputs a probability that a user will convert through an advertisement served to a webpage accessed on the user's computing device; a click-through model that outputs a probability that a user will click on an advertisement; a scroll interaction model that outputs a probability that a user will scroll back and forth over an advertisement at least a minimum number of times; a hotspot model that outputs a probability that a user will select at least a minimum number of hotspots within an interactive advertisement; a swipe model that outputs a probability that a user will swipe laterally through content within an advertisement; a virtual reality model that outputs a probability that a user will manipulate a virtual advertisement environment within an advertisement to at least a minimum degree; a video model that outputs a probability that a user will view at least a minimum duration or proportion of a video within an advertisement; and/or a brand lift model that outputs a probability that a user will exhibit at least a threshold increase in brand recognition after an advertisement is served to the user's computing device; etc.”).
With respect to Claim 8:
Narayan teaches:
The information processing apparatus according to claim 1, wherein the target information is an advertisement of the specific target (i.e. advertisement is of a specific type of product or brand) (Narayan: ¶ [0031] “The viewability model can also: intake metadata of an advertisement, such as the format of the advertisement ( e.g., static or interactive with video, catalog, virtual reality, or hotspot content) and a type of brand or product advertised; and output a probability that a user will scroll down to this advertisement inserted into a webpage viewed on the user's computing device and that the minimum proportion of this advertisement will be rendered on the user's computing device for at least the minimum duration of time based on historical user engagement data and these advertisement metadata.”).
With respect to Claim 9:
Narayan teaches:
The information processing apparatus according to claim 8, wherein the target information is a catch phrase of the specific target (i.e. textual statement pertaining to purchasing a product in the advertisement) (Narayan: ¶ [0025] “The visual element can include and/or animate a call to action (hereinafter "CTA''), such as a textual statement or icon configured to persuade a user to perform a particular task, such as purchasing a product, signing up for a newsletter, or clicking-through to a landing page for a brand or product.”).
With respect to Claim 10:
Narayan teaches:
The information processing apparatus according to claim 2, wherein the generation unit inputs information including the target information, the target impression information, the user impression information, and instruction information according to the improvement directionality information to generative AI as input information, and outputs the improvement information from the generative AI (i.e. artificial intelligence model utilizes correlations between advertisement information, user engagement, and predicted engagement in order to predict outcome and frequency brand lift of a particular product) (Narayan: ¶ [0113] “The remote computer system (or other computer system) can then implement linear regression, artificial intelligence, a convolutional neural network, or other analysis techniques to derive correlations between: engagement layer characteristics, mobile advertisement characteristics, user characteristics, and/or environment characteristics; and outcomes of composite mobile advertisements constructed from mobile/engagement layer pairs. The remote computer system can similarly derive correlations between these characteristics and outcomes of mobile advertisements served to users without engagement layers. For example, the remote computer system can identify: mobile advertisement format and engagement layer animation combinations that correlate with higher frequency instances of scroll events over an advertisement; engagement layers that correlate with higher frequency of conversions when placed in advertisements at the bottom of a webpage; and/or CTA placement and animations in an engagement layer that correlate with higher frequency of brand lift when paired with mobile advertisements advertising a particular category of product ( e.g., menswear, vehicles). The remote computer system ( or other computer system) can then generate an engagement layer model that represents these correlations, such as: one engagement layer model for each unique engagement layer hosted by the computer system; one engagement layer model representing predicted outcomes for multiple engagement layers applied to mobile advertisements within one advertising campaign; or one engagement layer model representing predicted outcomes for many engagement layers applied to mobile advertisements within any advertising campaign.”).
Claim(s) 13 is rejected under 35 U.S.C. 103 as being unpatentable over Narayan and Bradea in view of U.S. Patent 11,734,329 to Kershaw.
With respect to Claim 13:
Narayan does not explicitly disclose the information processing apparatus according to claim 1, wherein the estimation unit causes the generative artificial intelligence to generate, for each impression included in the impression group, reference information on which the user is estimated to have the impression, vectorizes the target information and the reference information, and compares similarity between the vectorized target information and the vectorized reference information to estimate the impression that the user is estimated to have.
However, Bradea further discloses wherein the estimation unit causes the generative artificial intelligence to generate, for each impression included in the impression group, reference information on which the user is estimated to have the impression, [[vectorizes the target information and the reference information, and compares similarity between the vectorized target information and the vectorized reference information to estimate the impression that the user is estimated to have]] (i.e. generative AI is trained to make predictions on data in order to determine outputs based on attributes and segment/event information of the user or estimate/infer attributes 102 using generative AI) (Bradea: ¶ [0028] “As used herein, the term "generative artificial intelligence (AI)" refers to hardware and/or software module(s) including program code that can create new content or make predictions based on the data it has been trained on. For example, a generative AI may include a chatbot that can accept arbitrary textual prompts and then generate novel responses. Generative AI systems can generate text, graphics, video, audio outputs, combinations thereof, or other output formats.” Furthermore, as cited in ¶ [0022] “The personalization module then determines a tuning parameter for a generative AI model, wherein the tuning parameter controls the randomness of the output of the generative AI model. For example, some generative AI models accept a temperature parameter for modulating the randomness of the output. A higher temperature corresponds to more randomness in the output given a prompt, where a lower temperature corresponds to less randomness given the same prompt. The personalization module may use information about the user interactions, including the events associated with those interactions, to determine the tuning parameter. For instance, the personalization module may use counts of events associated with the user or the product to determine the tuning parameter.” Furthermore, as cited in ¶¶ [0037] [0038] “For example, the content generation subsystem 130 may receive attributes 102, including information about the user and/or about the subject of the digital content ( e.g., a product)…The content generation subsystem 130 may also receive information about one or more segments to which the user belongs. For instance, the attributes 102 received by the content generation subsystem 130 may indicate that a web shopper is a 37 year-old female and that the subject of the content is a grey fleece. Identifying information about the user can be used to determine segments to which the user belongs. The example segment 104 describes its members as being Samsung users who enjoy mounting activities such as trekking. Segments may be determined on the basis of previously obtained information, retrieved from third-party sources or other databases, or inferred from user information and behaviors, among other possible mechanisms.”).
Therefore, it would have been obvious to one of ordinary skill in the art, at the time the invention was made, to add Bradea’s estimation unit causes the generative artificial intelligence to generate, for each impression included in the impression group, reference information on which the user is estimated to have the impression to Narayan’s target related information acquisition unit that acquires target information including information of a specific target; an estimation unit that estimates the impression that the user is estimated to have on the specific target from among a plurality of impressions included in an impression group; and a determination unit that determines the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group. One of ordinary skill in the art would have been motivated to do so because “Such precision marketing in near-real-time was not possible using existing tools, much less scalable. Thus, the techniques of the present disclosure can result in a higher conversion rate due to the highly personalized experience.” (Bradea: ¶ [0027]).
Narayan and Bradea do not explicitly disclose wherein the estimation unit [[causes the generative artificial intelligence]] to generate, for each impression included in the impression group, reference information on which the user is estimated to have the impression, vectorizes the target information and the reference information, and compares similarity between the vectorized target information and the vectorized reference information to estimate the impression that the user is estimated to have.
However, Kershaw further discloses wherein the estimation unit [[causes the generative artificial intelligence]] to generate, for each impression included in the impression group, reference information on which the user is estimated to have the impression, vectorizes the target information and the reference information, and compares similarity between the vectorized target information and the vectorized reference information to estimate the impression that the user is estimated to have (i.e. represent each chunk of text as a vector embedded in a high dimensional space representing semantic characteristics of the chunked text and categorize the chunked text into pre-defined categories using a threshold semantic similarity distance or hypersphere radius from any of a set of pre-defined anchor word sequences for each category) (Kershaw: Col. 9 Lines 3-26 “FIG. 3 is a diagram showing two different approaches to assigning an input embedded word sequence vector 313 (shown here as a point projected onto a2 dimensional plane however typically the vectors have several hundred dimensions) to a single pre-defined category "C" of interest using a semantic distance "hypersphere" approach. The black dots A1 311 and A2 321, represent anchor points also in a 2-dimensional projection 310, 320 of the high dimensional embedding vectors of the "anchor" text sequences used to pre-define a single category C. In the method on the left, the input embedded sequence vector 313 is assigned to the category C of interest if the distance from the input embedded sequence vector 313 to ANY anchor sequence vector 311, 321 is less than some threshold tightness distance "ri'' or "r2 " 312, 322 respectively for each anchor sequence vector 311, 321. In the figure FIG. 3 the input embedded sequence vector 313 falls within a threshold tightness distance r1 312 from anchor sequence vector A1 311 and so the input embedded sequence vector 313 is associated with category C, but this same input embedded sequence vector 313 falls outside the distance "r2" 322 from the anchor sequence vector A2 321 so it is not semantically close enough for A2 321 to be the cause of it being associated with the same category C.” Furthermore, as cited in Col. 13 Lines 3-20 “Vectorization of text 1320 is accomplished initially using a chunk-to-embedding sequence reducer 610 in a semantic similarity engine 420, which embeds a vector as metadata in a chunk, or alternatively this may be understood as generating additional data which is kept separate from the initial input text 405 and processed alongside it throughout the system, as metadata or "data about data." Categories are also determined after textual analysis 1330, through the use of a sequence embedder 620 which embeds sequences of words or a sequence of characters into metadata for semantic threshold analysis, and a semantic distance comparator 630 in a semantic similarity engine 420. These components draw correlations between categories and semantic vectors already determined about textual data, and compose or generate new categories and metadata about these relationships and a "bigger picture" view of the data, and do not have pre-determined or preset categories to choose from, unlike the operation of a deterministic rules engine 415.”).
Therefore, it would have been obvious to one of ordinary skill in the art, at the time the invention was made, to add Kershaw’s estimation unit to generate, for each impression included in the impression group, reference information on which the user is estimated to have the impression to Narayan’s target related information acquisition unit that acquires target information including information of a specific target; an estimation unit that estimates the impression that the user is estimated to have on the specific target from among a plurality of impressions included in an impression group; and a determination unit that determines the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group. One of ordinary skill in the art would have been motivated to do so in order for “allowing analysts or other users of the system to take action based on the results of the detailed sentiment and semantic analysis of the given data.” (Kershaw: Col. 12 Lines 61-63).
Claim(s) 6 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Narayan and Bradea in view of U.S. Patent 12,277,395 to Intrator.
With respect to Claim 6:
Narayan and Bradea do not explicitly disclose the information processing apparatus according to claim 5, wherein the determination unit determines an impression that the user has had on a basis of a posted message of the user with respect to the specific target.
However, Intrator further discloses wherein the determination unit determines an impression that the user has had on a basis of a posted message of the user with respect to the specific target (i.e. determine negative or positive impression on the basis of a posted message of the user) (Intrator: Col. 7 Lines 3-40 “After training and deployment of ML models 124 by SIP applications 120, one or more operations of SIP application 120 may utilize sentiment scores 126 with the post metadata from the metadata aggregator to generate a recommendation list. For example, sentiment scores 126 may be used when processing additional social media posts from social posts data 134 for identification of sentiments associated with an event or object (e.g., item, user, service entity, merchant, location, etc.) from the social media posts. Customer histories 116 may be used with the identified sentiments and posts, which may be processed and compared to a threshold number of posts and/or negative sentiments to determine whether there is a specific bad experience. Out of the box (00TB) solutions may be provided, such as when the negative sentiments may not indicate an overall bad experience, an experience not exceeding a threshold for identification of a particular sentiment, or of a neutral or other sentiment polarity. Using sentiment scores 126 with metadata and other information, recommended actions 128 may be determined. Recommended actions 128 may include actions to respond to a post and/or provide a comment, as well as actions to provide an additional social media post or communication to the customer, another associated customer or entity, and/or a public audience (e.g., available customers that may have or could have the same negative experience). Recommended actions 128 may also include one or more of providing additional computing services, identifying strategies or actions to assist with a good or bad experience and increase customer satisfaction and engagement, and/or recommending investigation into computing issues, downtime or interruption of service, fraud or risk, computing system attacks and vulnerabilities, poor customer service and/or experience issues, or the like. Thus, the social media post analytics, sentiment analysis for sentiment scores 126, and/or recommended actions 128 may be used with additional applications 112 to provide computing services, offers, and/or other engagements with customers and other users.”).
Therefore, it would have been obvious to one of ordinary skill in the art, at the time the invention was made, to add Intrator’s determination unit determines an impression that the user has had on a basis of a posted message of the user with respect to the specific target to Narayan’s target related information acquisition unit that acquires target information including information of a specific target; an estimation unit that estimates the impression that the user is estimated to have on the specific target from among a plurality of impressions included in an impression group; and a determination unit that determines the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group. One of ordinary skill in the art would have been motivated to do so in order “for a real-time scan and sentiment analysis of social media posts, including image posts and comments, using AI techniques, including NNs and other ML models, to more efficiently provide recommendations on the best course of action for a DCM to take.” (Intrator: Col. 2 Lines 35-40).
With respect to Claim 14:
Narayan and Bradea do not explicitly disclose the information processing apparatus according to claim 1, wherein the determination unit determines the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group on a basis of a posted message of the user with respect to the specific target.
However, Intrator further discloses wherein the determination unit determines the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group on a basis of a posted message of the user with respect to the specific target (i.e. determine negative or positive impression on the basis of a posted message of the user) (Intrator: Col. 7 Lines 3-40 “After training and deployment of ML models 124 by SIP applications 120, one or more operations of SIP application 120 may utilize sentiment scores 126 with the post metadata from the metadata aggregator to generate a recommendation list. For example, sentiment scores 126 may be used when processing additional social media posts from social posts data 134 for identification of sentiments associated with an event or object (e.g., item, user, service entity, merchant, location, etc.) from the social media posts. Customer histories 116 may be used with the identified sentiments and posts, which may be processed and compared to a threshold number of posts and/or negative sentiments to determine whether there is a specific bad experience. Out of the box (00TB) solutions may be provided, such as when the negative sentiments may not indicate an overall bad experience, an experience not exceeding a threshold for identification of a particular sentiment, or of a neutral or other sentiment polarity. Using sentiment scores 126 with metadata and other information, recommended actions 128 may be determined. Recommended actions 128 may include actions to respond to a post and/or provide a comment, as well as actions to provide an additional social media post or communication to the customer, another associated customer or entity, and/or a public audience (e.g., available customers that may have or could have the same negative experience). Recommended actions 128 may also include one or more of providing additional computing services, identifying strategies or actions to assist with a good or bad experience and increase customer satisfaction and engagement, and/or recommending investigation into computing issues, downtime or interruption of service, fraud or risk, computing system attacks and vulnerabilities, poor customer service and/or experience issues, or the like. Thus, the social media post analytics, sentiment analysis for sentiment scores 126, and/or recommended actions 128 may be used with additional applications 112 to provide computing services, offers, and/or other engagements with customers and other users.”).
Therefore, it would have been obvious to one of ordinary skill in the art, at the time the invention was made, to add Intrator’s determination unit determines the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group on a basis of a posted message of the user with respect to the specific target to Narayan’s target related information acquisition unit that acquires target information including information of a specific target; an estimation unit that estimates the impression that the user is estimated to have on the specific target from among a plurality of impressions included in an impression group; and a determination unit that determines the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group. One of ordinary skill in the art would have been motivated to do so in order “for a real-time scan and sentiment analysis of social media posts, including image posts and comments, using AI techniques, including NNs and other ML models, to more efficiently provide recommendations on the best course of action for a DCM to take.” (Intrator: Col. 2 Lines 35-40).
Claim(s) 15 and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Narayan and Bradea in view of U.S. Patent 10,395,258 to Akkiraju.
With respect to Claim 15:
Narayan and Bradea do not explicitly disclose the information processing apparatus according to claim 1, wherein the generation unit generates the prompt on a basis of template information corresponding to improvement directionality indicated by improvement directionality information, supplementary information including information indicating the specific target and a relationship between the result of the estimation and the result of the determination, and the merge information.
However, Akkiraju further discloses wherein the generation unit generates the prompt on a basis of template information corresponding to improvement directionality indicated by improvement directionality information, supplementary information including information indicating the specific target and a relationship between the result of the estimation and the result of the determination, and the merge information (i.e. templates are selected/composed according to a specified improvement goal/direction, the specified brand is the specific target, and the inferred (determined/actual) vs. intended (estimated/target) personality gap reads on the relationship) (Akkiraju: Example 3; Cols. 33-34 Lines 53-56 “Principle Factor Impacted: Marketing Message Imagery Description of Action: Send out messages that show more cheerfulness. Target: Clicking this link allows brand managers to provide their press release/marketing message and use IBM's Tone Analyzer™ service to refine the message to increase cheerfulness quantification. Benefit: A function that allows brand managers to assess what % of cheerfulness in what number of messages will lead to improvement in brand perception. The brand recommendation engine 960 may compose a combination of templates for different solutions based on a desired benefit so as to help a brand manager evaluate the benefits and decide on the right approach for perception improvement. For example, if the brand manager seeks to improve brand perception by 2%, the recommendation engine 960 may output a result that indicates that to improve brand perception by 2%, use solution template 1 to obtain 100 additional followers on Twitter™ and/or use solution template 2 to obtain 10 new employees and/or send out 5 marketing messages. The brand manager may then make an informed decision, based on an overlay of cost and time dimensions on the above functions, to find an optimal solution for achieving the desired benefit. Thus, the recommend solution is customized according to the specific brand comparison results 970. The recommended solution is output 990 for use by a brand manager or other authorized user. In one illustrative embodiment, the output 990 may be a notification of the actions that are recommended that the brand manager initiate to bring the inferred brand personality 920 closer to the intended brand personality 910. In other illustrative embodiments, the selected and customized solutions may include commands to be sent to other computing systems, applications, and the like, to initiate the actions recommended to improve the inferred brand personality 920 such that it more closely resembles that intended brand personality 910…FIG. 10 is a flowchart outlining an example operation for performing brand personality perception gap assessment in accordance with one illustrative embodiment. The operation outlined in FIG. 10 may be implemented, for example, by the brand personality perception gap assessment system 900 in FIG. 9. As shown in FIG. 10, the operation comprises receiving an intended brand personality (step 1010) and an inferred brand personality (step 1020). Gaps between the intended and inferred brand personalities are calculated (step 1030) and temporal changes of the inferred brand personality are calculated (step 1040). A corresponding output is generated that indicates the brand perception gaps and the temporal changes of the perceived personality of the brand (step 1050). The operation then terminates.”).
Therefore, it would have been obvious to one of ordinary skill in the art, at the time the invention was made, to add Akkiraju’s generation unit generates the prompt on a basis of template information corresponding to improvement directionality indicated by improvement directionality information, supplementary information including information indicating the specific target and a relationship between the result of the estimation and the result of the determination, and the merge information to Narayan’s target related information acquisition unit that acquires target information including information of a specific target; an estimation unit that estimates the impression that the user is estimated to have on the specific target from among a plurality of impressions included in an impression group; and a determination unit that determines the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group. One of ordinary skill in the art would have been motivated to do so in order “to analyze brands, determine divergent brand personalities from intended or desired brand personalities, generate recommendations for achieving the intended or desired brand personalities, and initiate, in some cases, the performance of actions to achieve the intended or desired brand personalities.” (Akkiraju: Col. 3 Lines 27-32).
With respect to Claim 16:
Narayan and Bradea do not explicitly disclose the information processing apparatus according to claim 1, wherein the generation unit generates impression information before and after improvement indicating a change in a degree of each impression that the user is estimated to have or the user is determined to have had before and after improvement by the improvement information.
However, Akkiraju further discloses wherein the generation unit generates impression information before and after improvement indicating a change in a degree of each impression that the user is estimated to have or the user is determined to have had before and after improvement by the improvement information (i.e. generate impression information including before or inferred state/current brand personality and an after improvement or after 2% improvement change) (Akkiraju: Example 3; Cols. 33-34 Lines 53-9 “Principle Factor Impacted: Marketing Message Imagery Description of Action: Send out messages that show more cheerfulness. Target: Clicking this link allows brand managers to provide their press release/marketing message and use IBM's Tone Analyzer™ service to refine the message to increase cheerfulness quantification. Benefit: A function that allows brand managers to assess what % of cheerfulness in what number of messages will lead to improvement in brand perception. The brand recommendation engine 960 may compose a combination of templates for different solutions based on a desired benefit so as to help a brand manager evaluate the benefits and decide on the right approach for perception improvement. For example, if the brand manager seeks to improve brand perception by 2%, the recommendation engine 960 may output a result that indicates that to improve brand perception by 2%, use solution template 1 to obtain 100 additional followers on Twitter™ and/or use solution template 2 to obtain 10 new employees and/or send out 5 marketing messages. The brand manager may then make an informed decision, based on an overlay of cost and time dimensions on the above functions, to find an optimal solution for achieving the desired benefit.”).
Therefore, it would have been obvious to one of ordinary skill in the art, at the time the invention was made, to add Akkiraju’s generation unit generates impression information before and after improvement indicating a change in a degree of each impression that the user is estimated to have or the user is determined to have had before and after improvement by the improvement information to Narayan’s target related information acquisition unit that acquires target information including information of a specific target; an estimation unit that estimates the impression that the user is estimated to have on the specific target from among a plurality of impressions included in an impression group; and a determination unit that determines the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group. One of ordinary skill in the art would have been motivated to do so in order “to analyze brands, determine divergent brand personalities from intended or desired brand personalities, generate recommendations for achieving the intended or desired brand personalities, and initiate, in some cases, the performance of actions to achieve the intended or desired brand personalities.” (Akkiraju: Col. 3 Lines 27-32).
Response to Arguments
Applicant’s arguments see page 8 of the Remarks disclosed, filed on 06/30/2026, with respect to the 35 U.S.C. § 112(f) interpretation of claim(s) 1, 2, 5, and 7 have been considered and are persuasive. The Applicant asserts “Applicant acknowledges the Examiner's interpretation of the claim terms "target related information acquisition unit," "generation unit," "reception unit," "determination unit," and "estimation unit" under 35 U.S.C. § 112(f). Applicant does not dispute the Examiner's interpretation at this time and confirms that the corresponding structure for these claim terms is the processing unit hardware described in the specification as published at paragraphs [0126]- [0127], which describes implementation by a processor such as a CPU or MPU, or by an integrated circuit such as an ASIC, FPGA, or GPGPU. Applicant further notes that amended claim 1 now recites "at least one memory" and "at least one processor coupled to the at least one memory," which provides explicit structural elements that address the Examiner's § 112(f) interpretation.” The Examiner agrees and therefore, the interpretation of claim(s) 1, 2, 5, and 7 under 35 U.S.C. § 112(f) has been withdrawn.
Applicant’s arguments see pages 8-9 of the Remarks disclosed, filed on 06/30/2026, with respect to the 35 U.S.C. § 101 rejection(s) of claim(s) 1-12 have been considered but are not persuasive:
The Applicant asserts “As an initial matter, the amended claims are not directed to an abstract idea under Step 2A, Prong I. Amended claim 1 now recites a specific technical process comprising: (a) an estimation unit that estimates the impression that the user is estimated to have on the specific target from among a plurality of impressions included in an impression group using generative artificial intelligence on a basis of the target information; (b) a determination unit that determines the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group using the generative artificial intelligence on a basis of user action information indicating an action of the user regarding the specific target; and (c) a generation unit that generates merge information by merging a result of the estimation by the estimation unit and a result of the determination by the determination unit, generates a prompt on a basis of the merge information, and inputs the prompt to the generative artificial intelligence to cause the generative artificial intelligence to output improvement information. This is not a mental process; a human cannot practically estimate and determine impressions from a predefined impression group using generative artificial intelligence, merge those results into merge information, construct a prompt from the merge information, and feed the prompt to generative artificial intelligence to generate improvement content. The claims recite a specific, computer- implemented technical process that is fundamentally different from the mental and organizational activities identified by the Examiner. Moreover, amended claim 1 now recites explicit hardware, i.e., at least one processor and at least one memory, further distinguishing the claims from an abstract mental process.” The Examiner respectfully disagrees. Claims 1, 11, and 12 recite limitations directed to the abstract idea including “acquiring target information including information of a specific target and target; estimating the impression that the user is estimated to have on the specific target from among a plurality of impressions included in an impression group; determining the impression that the user is determined to have had on the specific target from among the plurality of impressions included in the impression group; generating merge information by merging a result of the estimating and a result of the determining; and generating a prompt on a basis of the merge information.” These further limitations are not seen as any more than the judicial exception. Claims 1, 11, and 12 recite additional limitations including “using generative artificial intelligence on a basis of the target information and/or user action information indicating an action of the user regarding the specific target; and inputting the prompt to the generative artificial intelligence to cause the generative artificial intelligence to output improvement information including information indicating improvement content related to the target information.” The claims are considered to be an abstract idea under certain methods of organizing human activity because the claims are directed to commercial or legal interactions (including agreements in the form of contracts; legal obligations; advertising, marketing or sales activities or behaviors; business relations) and managing personal behavior or relationships or interactions between people (including social activities, teaching, and following rules or instructions) such as improving content based on estimated and determined information. The claims are also considered to be an abstract idea under Mental Processes such as concepts performed in the human mind (including an observation, evaluation, judgment, opinion) because the claims are directed to acquiring information (i.e. target information, target impression information, and user impression information), estimating information (i.e. estimating user impression), determining information (i.e. if the impression occurred), and generating information (i.e. merge information).
The Applicant also asserts “Even if the claims were found to recite an abstract idea, they integrate it into a practical application under Step 2A, Prong II. Amended claim 1 recites a specific technological implementation: using generative artificial intelligence to classify target information and user action information into impressions from a predefined impression group, merging those classification results into merge information, constructing a prompt based on the merge information, and feeding the prompt to generative artificial intelligence to generate improvement content. This is not merely "apply it". Rather, it is a specific, ordered combination of technical steps that produces a concrete result (improvement information for target content). The use of generative artificial intelligence here is not as a generic tool but as an integral part of a multi-step technical process involving impression classification from an impression group, result merging, and prompt-based content generation.” The Examiner respectfully disagrees. The claim does not recite any technical detail about how the generative AI performs the classification into “a plurality of impressions included in an impression group” — no model architecture, no specific training methodology, no particular data representation or algorithm is claimed. Reciting a result (“cause the generative artificial intelligence to output improvement information”) is result-oriented functional claiming, not a technical implementation. See also MPEP § 2106.05(f) (mere instructions to “apply” an abstract idea using a generic technology adjunct does not integrate the idea into a practical application). Furthermore, Claims 1, 11, and 12 recite additional limitations including “using generative artificial intelligence on a basis of the target information and/or user action information indicating an action of the user regarding the specific target; and inputting the prompt to the generative artificial intelligence to cause the generative artificial intelligence to output improvement information including information indicating improvement content related to the target information.” “Using generative artificial intelligence” appears twice, both times as a black-box tool invoked to perform the abstract mental step (estimate an impression; determine an impression) — no technical detail is given about how the generative AI accomplishes this (no model architecture, no training method, no specific data representation). “Generating a prompt on a basis of the merge information” and “inputting the prompt to the generative artificial intelligence” is simply describing the generic, well-known mechanics of using an LLM — assemble text, send it to the model, get text back. This is precisely the kind of generic, off-the-shelf, functionally-claimed use of AI that recent case law has found insufficient to integrate an abstract idea into a practical application. These additional limitations are seen as adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f). Accordingly, alone, and in combination, these additional elements are seen as using a computer or tool to perform an abstract idea, adding insignificant-extra-solution activity to the judicial exception. They do no more than link the judicial exception to a particular technological environment or field of use (i.e. generative AI) and therefore do not integrate the abstract idea into a practical application. Recentive Analytics, Inc. v. Fox Corp., 134 F.4th 1205 (Fed. Cir. 2025) — the Federal Circuit held claims reciting the use of machine learning to generate network maps and event schedules were ineligible, because the claims merely applied generic, unspecified machine learning techniques to a new problem/data domain, without claiming any improvement to how the machine learning model itself works. The court emphasized that “training a machine learning model” and “applying” it to generate an output, without more, does not transform an abstract idea into a patent-eligible invention — result-oriented, functional claiming of AI is not itself an inventive technical solution. This claim has the same defect: it claims the result (“cause the generative artificial intelligence to output improvement information”) without claiming how the generative AI is technically adapted, improved, or configured to achieve that result better than any generic LLM would if simply prompted with the same merged data. There’s no claimed improvement to: the underlying AI model architecture, the computer’s functioning, a technical field, or any particular machine or transformation of matter. The courts decided that although the additional elements did limit the use of the abstract idea, the court explained that this type of limitation merely confines the use of the abstract idea to a particular technological environment and this fails to add an inventive concept to the claims (See Affinity Labs of Texas v. DirecTV, LLC,). Therefore, the rejection(s) of claim(s) 1-16 under 35 U.S.C. § 101 is maintained above with an updated analysis.
Applicant’s arguments see pages 10-11 of the Remarks disclosed, filed on 06/30/2026, with respect to the 35 U.S.C. § 102(a)(1) rejection(s) of claim(s) 1-5 and 7-12 over Narayan have been considered but are moot because the arguments do not apply to the new ground(s) of rejection is made in view of U.S. Publication 2025/0068893 to Bradea.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. The following references are cited to further show the state of the art:
U.S. Publication 2014/0337120 to Ercanbrack for disclosing An apparatus, system, and method are disclosed for integrating media analytics to configure an advertising engine. In one embodiment, a method includes advertising based on an advertising campaign, the advertising campaign comprising a plurality of advertising entities, the advertising campaign based on a set of configuration parameters, monitoring results of one or more of the advertising entities, and modifying one or more of the configuration parameters based on the results.
U.S. Publication 2018/0308124 to Gao for disclosing Machine learning techniques are described for generating recommendations using decision trees. A decision tree is generated based on training data that comprises multiple training instances, each of which comprises a feature value for each of multiple features and a label of a target variable. The multiple features correspond to attributes of multiple content delivery campaigns. Later, feature values of a content delivery campaign are received. The decision tree is traversed using the feature values to generate output. Based on the output, one or more recommendations are identified and the one or more recommendations are presented on a computing device.
U.S. Publication 2025/0299672 to Suzuki for disclosing A determination device according to one aspect according to the present disclosure includes an input unit that receives an input of reaction information indicating a reaction of a user, a generation unit that generates first persona information that is information generated on the basis of the reaction information received by the input unit, the information indicating a characteristic of the user, a determination unit that determines consistency between the first persona information generated by the generation unit and second persona information based on past reaction information of the user, and an update unit that updates the persona information regarding the user on the basis of the consistency determined by the determination unit.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, 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 Azam Ansari, whose telephone number is (571) 272-7047. The examiner can normally be reached from Monday to Friday between 8 AM and 4:30 PM.
If any attempt to reach the examiner by telephone is unsuccessful, the examiner's supervisor, Waseem Ashraf, can be reached at (571) 270-3948.
Another resource that is available to applicants is the Patent Application Information Retrieval (PAIR). Information regarding the status of an application can be obtained from the (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAX. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pairdirect.uspto.gov. Should you have questions on access to the Private PAIR system, please feel free to contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
Applicants are invited to contact the Office to schedule either an in-person or a telephonic interview to discuss and resolve the issues set forth in this Office Action. Although an interview is not required, the Office believes that an interview can be of use to resolve any issues related to a patent application in an efficient and prompt manner.
/AZAM A ANSARI/
Primary Examiner, Art Unit 3621
September 5, 2026
Prosecution Insights