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 .
Claim Objections
Claim 1 is/are objected to because of the following informalities:
Claim 1: “based on the determined at least one the price per follower or the adjusted price per follower” should read “based on the determined at least one of the price per follower or the adjusted price per follower”
Appropriate correction is required.
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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Claim(s) 1/20 is/are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Claim(s) 1/20 is/are directed towards a computer system (i.e. machine) and a method (i.e. a process), respectively. Thus, each of the claims fall within one of the four statutory categories. Nevertheless, the claims fall within the judicial exception of an abstract idea.
Claim(s) 1/20 is/are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claim recites “receive real market data, the real market data including completed deal data and disclosure data; receive the user input data; retrieve a real-time current follower count for the user using the received user channel identifier data; filter, using the valuation model, the received real market data based on the received user input data to generate a filtered dataset; determine, via the valuation model, at least one of a price per follower or an adjusted price per follower based on the retrieved real-time current follower count; generate an adjusted dataset, using the valuation model, by adjusting the filtered received real market data based on the determined at least one the price per follower or the adjusted price per follower; generate one or more match level tables, using the valuation model, by reducing the adjusted dataset based on one or more predetermined thresholds and the automated parameter tuning; generate a final dataset based on the generated one or more match level tables using the valuation model; and determine a suggested activity price for the user, using the valuation model, based on the generated final dataset.”
The limitations above, as drafted, is a process that, under its broadest reasonable interpretation, covers a method of “determining a suggested activity price by collecting market data, filtering comparable transactions and applying pricing factors and outputting a suggested price” which is a method of organizing a human activity, mental process and mathematical concepts. That is, the method allows for fundamental economic principles or practices (including hedging, insurance, mitigating risk); commercial or legal interactions (including agreements in the form of contracts; legal obligations; advertising, marketing or sales activities or behaviors; business relations); managing personal behavior or relationships or interactions between people (including social activities, teaching, and following rules or instructions); concepts performed in the human mind and mathematical relationships/formula/equations/calculations.
This judicial exception is not integrated into a practical application. In particular, the claim recites “a user interface device including a display and a user input device, the user input device configured to receive user input data from a user via the user input device, the user input data including at least activity type data, user identifier data, and user channel identifier data”, “a platform server including one or more processors configured to execute a set of program instructions stored in a memory, the platform server including a valuation model stored in the memory, wherein the valuation model includes a trained machine learning classifier, wherein the platform server is communicatively coupled to the user interface device via a network”, “a database” and “perform, using the trained machine learning classifier of the valuation model, automated parameter tuning on the adjusted dataset based on one or more dynamic parameters, wherein the one or more dynamic parameters include at least one of a dataset size parameter, a log denominator parameter, a weight parameter, a share parameter, or a decay parameter” (claim 1); “database”, user input device”, “performing, using a trained machine learning classifier of the valuation model, automated parameter tuning on the adjusted dataset based on one or more dynamic parameters, wherein the one or more dynamic parameters include at least one of a dataset size parameter, a log denominator parameter, a weight parameter, a share parameter, or a decay parameter” (claim 20). Each of the additional limitations is recited at a high level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component. Accordingly, these additional element(s), alone or in combination, do(es) not integrate the abstract idea into a practical application because it/they do(es) not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea.
The claim does not include additional elements that are sufficient to 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 element(s) is/are nothing more than mere instructions to apply the exception on a general computer.
Dependent claim(s) 12 is/are also directed to an abstract idea without significantly more because it/they further narrow(s) the abstract idea described in relation to claim 1 without successfully integrating the exception into a practical application (“calculating…,storing… and selecting… are recited at a high level of recitation which amounts to mere instructions to apply the exception in a computer environment because they are generic mathematical/model optimization operations used to improve the accuracy of the abstract calculation of the suggested price) or providing significantly more limitations.
Dependent claim(s) 2-11, 13, 16 and 18-19 is/are also directed to an abstract idea without significantly more because it/they further narrow(s) the abstract idea described in relation to claim 1 without successfully integrating the exception into a practical application or providing significantly more limitations.
Dependent claim(s) 14 is/are also directed to an abstract idea without significantly more because it/they further narrow(s) the abstract idea described in relation to claim 1 without successfully integrating the exception into a practical application (social media channel handle or social media channel profile link is recited at a high level of recitation which amounts to mere instructions to apply the exception in a computer environment because they are generic mathematical/model optimization operations used to improve the accuracy of the abstract calculation of the suggested price) or providing significantly more limitations.
Dependent claim(s) 15 is/are also directed to an abstract idea without significantly more because it/they further narrow(s) the abstract idea described in relation to claim 1 without successfully integrating the exception into a practical application (“generate one or more Application Programming Interface requests for one or more social media platform servers based on the received social media channel handle or the received social media channel profile link; and retrieve the real-time current follower count for the user based on the generated one or more Application Programming Interface requests” is recited at a high level of recitation which amounts to mere instructions to apply the exception in a computer environment because Application Programming interface and using them to retrieve data are generic computer components/function) or providing significantly more limitations.
Dependent claim(s) 17 is/are also directed to an abstract idea without significantly more because it/they further narrow(s) the abstract idea described in relation to claim 1 without successfully integrating the exception into a practical application (“generate one or more control signals configured to cause the display of the user device to display the determined suggested activity price” is recited at a high level of recitation which amounts to mere instructions to apply the exception in a computer environment) or providing significantly more limitations.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 1-2, 4-9, and 13-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Lawrence (US 20230016916) in view of Walters (US 20210264199) and Sullivan (US 20190012746).
As per claim 1/20, Lawrence discloses a system, the system comprising:
a user interface device including a display and a user input device, the user input device configured to receive user input data from a user via the user input device, the user input data including at least activity type data, user identifier data, and user channel identifier data (“[0024] In embodiment, the one or more user devices 110 may be configured to receive one or more user inputs from a user. For example, the one or more user devices 110 may include a user interface, wherein the user interface includes a display 114 and a user input device 116. The one or more processors 104 may be configured to generate the graphical user interface of the display 114, wherein the graphical user interface includes the one or more display pages configured to transmit and receive data to and from a user.”, “[0004]… the user input data including at least activity type data, user identifier data, and user channel identifier data”); and
a platform server including one or more processors configured to execute a set of program instructions stored in a memory, the platform server including a valuation model stored in the memory, wherein the valuation model, wherein the platform server is communicatively coupled to the user interface device via a network, wherein the set of program instructions are configured to cause the one or more processors to (“[0020] In embodiments, the system 100 includes one or more platform servers 102. The one or more platform servers 102 may include one or more processors 104 configured to execute program instructions maintained on a memory medium 106. In this regard, the one or more processors 104 of the one or more platform servers 102 may execute any of the various process steps described throughout the present disclosure. For example, the one or more processors 104 may be configured to determine activity pricing for a user (e.g., student athlete, professional athlete, coach, or the like) based on a valuation model 108 stored in memory 106. The valuation model 108 may use real market data corresponding to that individual's unique characteristics (e.g., gender, sport, position, institution, conference, number of follower, and the like) to calculate a suggested activity pricing. In this regard, the activity pricing may be beneficial in evaluating whether a sponsorship deal is appropriate. Further, the one or more platform servers 102 may be configured to receive data including, but not limited to, real market data, user data, and the like.”, “[0021] In embodiments, the one or more platform servers 102 may be communicatively coupled to one or more user devices 110 via the network 112. For example, the one or more platform servers 102 and/or the one or more user devices 110 may include a network interface device and/or the communication circuitry suitable for interfacing with the network 112.”):
receive real market data from a database, the real market data including completed deal data and disclosure data ([0032] In step 202, the system 100 may receive real market data. For example, the one or more processors 104 of the platform server 102 may be configured to receive real market data from a database 118 (stored in memory 106 or a remote database) to train the valuation model 108 stored in memory 106. The database 118 may include real market data such as, but is not limited to, completed deals (e.g., deals completed using the platform server and stored in the platform database), disclosures (e.g., disclosed deals performed by individuals off the platform), or the like.”);
receive the user input data from the user input device (“[0004]… the system includes a user interface device including a display and a user input device, the user device configured to receive user input data from a user via the user input device, the user input data including at least activity type data, user identifier data, and user channel identifier data. “);
retrieve a real-time current follower count for the user using the received user channel identifier data (0039] In Eqn. 1, the activity price may be the suggested activity price (calculated in step 220). The one or more processors 104 of the platform server 102 may be configured to determine a real-time follower count based the user's inputted social media handle or profile link. For example, the user may input their social media handle or profile link such that the one or more processors 104 of the platform server 102 may be able to retrieve the user's real-time follower count.);
filter, using the valuation model, the received real market data based on the received user input data to generate a filtered dataset (“[0037] In step 206, the system 100 may filter the received real market data based on the received user data. In one non-limiting example, the one or more processors 104 of the platform server 102 may be configured to filter the received real market data, via the valuation model 108, based at least one of a selected identifier (e.g., which sport an individual participates in) or a selected activity type received from the user (in step 204). In this example, the one or more processors 104 of the platform server 102 may be configured to filter the received real market data based on the student athlete identifier and social post activity type. In this regard, the calculated activity pricing (calculated in step 220) may provide an accurate estimate of a user's market value for a specific social post activity type based on relevant real market data corresponding to the student athlete market. For example, in a non-limiting example, if a Division I quarterback does an Instagram post for $2,000, then the valuation model 108 may be configured to determine what an accurate suggested activity price should be for a similar individual and similar activity type based on the received real market data.”);
determine, via the valuation model, at least one of a price per follower or an adjusted price per follower based on the retrieved real-time current follower count ([0038] In an optional step 208, if social media follower count is known, the system 100 may determine an activity price per follower (PPF). For example, the one or more processors 104 of the platform server 102 may be configured to determine an activity PPF, using the valuation model 108, based on Equation 1 (Eqn. 1), [0040] In an optional step 210, if social media follower count is known, the system 100 may determine an adjusted PPF. For example, the one or more processors 104 of the platform server 102 may be configured to determine an adjusted PPF, using the valuation model 108, based on Equation 2 (Eqn. 2), which is shown and described below:);
generate an adjusted dataset, using the valuation model, by adjusting the filtered received real market data based on the determined at least one the price per follower or the adjusted price per follower ([0043] In step 214, the system 100 may generate an adjusted dataset based on at least one of the calculated PPF (step 208), adjusted PPF (step 210), or activity price (step 212). For example, the adjusted dataset may be weighted by buyer type, such that the non-fan buyer would be discounted compared to a fan.);
one or more dynamic parameters, wherein the one or more dynamic parameters include at least one of a dataset size parameter, a log denominator parameter, a weight parameter, a share parameter, or a decay parameter ([0047] In step 218, the system 100 may generate a final dataset. For example, the one or more processors 104 of the platform server 102, using the valuation model 108, may be configured to generate a final dataset based on the generated match table (in step 216) by duplicating the number of times the user input data matches the data in the match level table. For instance, the one or more processors 104 of the platform server 102 may be configured to generate a final dataset, where the match level table is sorted by match level (ascending) and activity date (descending). In a non-limiting example, the top 100 rows/activities of the match level table may be kept. Further, 25% of the dataset may be reserved for market influence (e.g., excluding match level 1) to prevent an athlete who has done a lot of deals from going stale if the market spikes. It is noted that the final dataset may include any amount of comparison data (e.g., rows of data) suitable for determining the suggested activity price (in step 220).”);
generate one or more match level tables, using the valuation model, by reducing the adjusted dataset based on one or more predetermined thresholds and parameters (0045] For example, as shown in Table 2, a match table may be generated based one or more predetermined thresholds associated with one or more match levels. In one instance, a first portion of the match table may be generated for a match level 1 including data that matches the “exact athlete”, where there may be 25 datapoints (or duplications). In another instance, a second portion of the match table may be generated for a match level 2 including data that matches the “sport+institution”, where there may be 15 datapoints (or duplications). In another instance, a third portion of the match table may be generated for a match level 3 including data that matches the “sport+conference”, where there may be 10 datapoints (or duplications). In another instance, a fourth portion of the match table may be generated for a match level 4 including data that matches the “sport+league/division”, where there may be 5 datapoints (or duplications). In another instance, a fifth portion of the match table may be generated for a match level 5 including data that matches the “institution”, where there may be 3 datapoints (or duplications). In another instance, a sixth portion of the match table may be generated for a match level 6 including data that matches the “conference”, where there may be 2 datapoints (or duplications). In another instance, a seventh portion of the match table may be generated for a match level 7 including data that matches the “league/division”, where there may be 1 datapoint (or duplications)…[0047] In step 218, the system 100 may generate a final dataset. For example, the one or more processors 104 of the platform server 102, using the valuation model 108, may be configured to generate a final dataset based on the generated match table (in step 216) by duplicating the number of times the user input data matches the data in the match level table. For instance, the one or more processors 104 of the platform server 102 may be configured to generate a final dataset, where the match level table is sorted by match level (ascending) and activity date (descending). In a non-limiting example, the top 100 rows/activities of the match level table may be kept. Further, 25% of the dataset may be reserved for market influence (e.g., excluding match level 1) to prevent an athlete who has done a lot of deals from going stale if the market spikes. It is noted that the final dataset may include any amount of comparison data (e.g., rows of data) suitable for determining the suggested activity price (in step 220).”) ;
generate a final dataset based on the generated one or more match level tables using the valuation model ([0047] In step 218, the system 100 may generate a final dataset. For example, the one or more processors 104 of the platform server 102, using the valuation model 108, may be configured to generate a final dataset based on the generated match table (in step 216) by duplicating the number of times the user input data matches the data in the match level table. For instance, the one or more processors 104 of the platform server 102 may be configured to generate a final dataset, where the match level table is sorted by match level (ascending) and activity date (descending). In a non-limiting example, the top 100 rows/activities of the match level table may be kept. Further, 25% of the dataset may be reserved for market influence (e.g., excluding match level 1) to prevent an athlete who has done a lot of deals from going stale if the market spikes. It is noted that the final dataset may include any amount of comparison data (e.g., rows of data) suitable for determining the suggested activity price (in step 220).”); and
determine a suggested activity price for the user, using the valuation model, based on the generated final dataset ([0048] In a step 220, the system 100 may determine a suggested activity price. For example, the one or more processors 104 of the platform server 102 may be configured to determine a suggested activity price, using the valuation model 108, based on Equation 3 (Eqn. 3), which is shown and described below:… [0049] For instance, the one or more processors 104 of the platform server 102 may be configured to determine the suggested activity price based on the follower count received from the user (in step 204) and the calculated adjusted PPF (in step 210), where the one or more processors 104 of the platform 102 may be configured to determine the mean value of the calculated adjusted PPF (from step 210).”),
However, Lawrence does not disclose but Walters discloses perform, using the trained machine learning classifier of the valuation model, automated parameter tuning on the adjusted dataset based on hyperparameters (“[0013] Embodiments disclosed herein use machine learning to control the tuning of hyperparameters of an AI model specified to be used to perform a particular function. Generally, as the tuning of hyperparameters for the AI model begins, evaluations of the results of initial iterations of such tuning may be used to train one or more prediction models. During subsequent iterations of such tuning, the one or more prediction models may then be used to generate predictions concerning the efficacy of subsequent iterations of such tuning as part of determining when to cease such tuning.”…” [0022] FIG. 1 depicts a schematic of an exemplary system 100 for the tuning of hyperparameters of an AI model, consistent with disclosed embodiments. As shown, the system 100 may include a requesting device 102, one or more data devices 103, a tuning device 104, and/or one or more node devices 105. The requesting device 102 may provide the tuning device 104 with request data 234 conveying details of a request to tune the hyperparameters of an AI model. The one or more data devices 103 may provide the tuning device 104 with a training data and/or testing data for use in such tuning… [0059] Turning to FIG. 6D, regardless of whether the processing and/or storage resources of the tuning device 104 are used to perform the tuning of hyperparameters of the AI model, or the processing and/or storage resources of the one or more node devices 105 are so used, following the testing of the batch 670 of instances 673 of the AI model by either of the testing components 445 or 545, the hyperparameter generation component 442 may employ indications of the results of such testing to guide its generation of a next batch 630 of sets 632 of hyperparameters.“
Therefore, it would have been obvious to one of ordinary skill in the art at the time of the invention to include the limitations as taught by Walters in the teaching of Lawrence, in order to estimate whether further application of a hyperparameter tuning technique will cause an improvement in at least one of the hyperparameters (please see Walters abstract).
However, Lawrence does not disclose expressly disclose that the valuation model includes a trained machine learning classifier but Sullivan discloses using a trained machine learning classifier, including a random forest classifier to classify influencers based on a social media and campaign related data ([0014] With that being said, classification is a statistical process used to partition a collection of items (e.g., social media network metrics) into homogeneous classes according to their measurable characteristics or features. Generally, a typical classifier is first trained via machine learning techniques to recognize and label key patterns in a set of available training samples, and is then used to predict the class membership of future data…. [0015] One of the most recent advances in classification is the random forest (RF) methodology, which is a non-parametric ensemble approach to machine learning that uses bagging to combine the decisions of multiple classification trees to classify data samples. Of the many classifiers that have been developed, few have addressed the aforementioned issues as effectively as the RF, which has been demonstrated to be highly accurate, robust, easy to use, and resistant to overtraining…. [0018] As such, a need exists for a system and method utilizing RF methodology to identify the best social media influencer for a specific campaign utilizing key performance indicators to judge a specific influencer's influencing power as well as metrics specific to the business objectives of the individual or company wishing to enlist the services of a social media influencer.)
Therefore, it would have been obvious to one of ordinary skill in the art at the time of the invention to include the limitations as taught by Sullivan in the teaching of Lawrence, in order to identify the best social media influencer for a specific campaign (please see Sullivan paragraph 18).
As per claim 2, Lawrence discloses wherein the dataset size parameter defines a maximum threshold for a number of activities used to determine the suggested activity price, wherein the one or more processors are configured to determine the suggested activity price based on a predetermined number of most recent activities of the real market data based on the minimum threshold and the maximum threshold ([0047] In step 218, the system 100 may generate a final dataset. For example, the one or more processors 104 of the platform server 102, using the valuation model 108, may be configured to generate a final dataset based on the generated match table (in step 216) by duplicating the number of times the user input data matches the data in the match level table. For instance, the one or more processors 104 of the platform server 102 may be configured to generate a final dataset, where the match level table is sorted by match level (ascending) and activity date (descending). In a non-limiting example, the top 100 rows/activities of the match level table may be kept. Further, 25% of the dataset may be reserved for market influence (e.g., excluding match level 1) to prevent an athlete who has done a lot of deals from going stale if the market spikes. It is noted that the final dataset may include any amount of comparison data (e.g., rows of data) suitable for determining the suggested activity price (in step 220). ).
However, while Lawrence discloses maximum threshold for the data set it does not explicitly disclose a minimum threshold for the data set but Walters discloses that a minimum threshold for data size can be set (paragraph 30)(please see claim 1 rejection for combination rationale).
As per claim 4, Lawrence discloses wherein the weight parameter defines a relative importance of activities at each match level within the adjusted dataset by determining a number of duplications for an activity at a given match level relative to a total dataset size of the final dataset generated ([0045] For example, as shown in Table 2, a match table may be generated based one or more predetermined thresholds associated with one or more match levels. In one instance, a first portion of the match table may be generated for a match level 1 including data that matches the “exact athlete”, where there may be 25 datapoints (or duplications). In another instance, a second portion of the match table may be generated for a match level 2 including data that matches the “sport+institution”, where there may be 15 datapoints (or duplications). In another instance, a third portion of the match table may be generated for a match level 3 including data that matches the “sport+conference”, where there may be 10 datapoints (or duplications). In another instance, a fourth portion of the match table may be generated for a match level 4 including data that matches the “sport+league/division”, where there may be 5 datapoints (or duplications). In another instance, a fifth portion of the match table may be generated for a match level 5 including data that matches the “institution”, where there may be 3 datapoints (or duplications). In another instance, a sixth portion of the match table may be generated for a match level 6 including data that matches the “conference”, where there may be 2 datapoints (or duplications). In another instance, a seventh portion of the match table may be generated for a match level 7 including data that matches the “league/division”, where there may be 1 datapoint (or duplications)… [0047] In step 218, the system 100 may generate a final dataset. For example, the one or more processors 104 of the platform server 102, using the valuation model 108, may be configured to generate a final dataset based on the generated match table (in step 216) by duplicating the number of times the user input data matches the data in the match level table.).
As per claim 5, Lawrence discloses wherein the share parameter includes a maximum cumulative percentage of the adjusted dataset that a match level and all previous match levels collectively occupied to diversify the final dataset and prevent any single match level from dominating pricing calculations of the determined suggested activity price ([0047]… Further, 25% of the dataset may be reserved for market influence (e.g., excluding match level 1) to prevent an athlete who has done a lot of deals from going stale if the market spikes. It is noted that the final dataset may include any amount of comparison data (e.g., rows of data) suitable for determining the suggested activity price (in step 220). Under BRI, this 25% market influence reservation functions as a maximum cumulative percentage cap on match level 1, by reserving 25% exclusively for market influence data ecluding match level 1, the system caps match level 1’s max share of the final dataset at 75%, directly preventing match level 1 from dominating pricing calculations. Lawrence expressly states the purpose of this reservation is to “prevent an athlete who has done a lot of deals from going stale if the market spikes” (i.e. to diversify the final dataset and prevent a single match level from dominating pricing).
As per claim 6, Lawrence discloses wherein the decay parameter represents a depreciation of activity value as matching criteria associated with the generated match level becomes less precise across different match levels (paragraph 45 discloses a seven level match hierarchy in table 2 where the duplication count decreases monotonically as match precision decreases from 25 duplications for the most precise match down to 1 duplication for the least precise match. Under BRI, this decreasing duplication count directly corresponds to depreciation of activity value as matching criteria becomes less precise. Because the suggested activity price is determined as the mean adjusted PPF across the final dataset, each activity’s contribution to that mean is proportional to its duplication count. An activity at match level 7 contributes 1/25th the pricing weight of an equivalent activity at match level 1, a quantified depreciation of its effective value in the pricing calculation that is directly tied to the reduced precision of its matching criteria).
As per claim 7, Lawrence discloses wherein the one or more processors are configured to: determine a buyer type modifier based on a historical buyer spend amount ([0041] The buyer modifier may include a donor modifier, sponsor modifier, brand modifier, fan modifier, a collective modifier (e.g., specific group of individuals who support a particular institution), and the like. In one non-limiting example, the modifiers may be 0.10 for a donor, 0.50 for a sponsor, 0.75 for a brand, and 1.00 for a fan. In another non-limiting example, the modifiers may be 0.10 for a donor, 0.15 for a sponsor, 0.20 for a brand, and 1.00 for a fan. In another non-limiting example, the modifiers may be 0.10 for a donor, 0.15 for a sponsor, 0.20 for a brand, 0.50 for a collective, and 1.00 for a fan. It is noted that the buyer modifier may be any predetermined modifier factor configured to weight the value.” These modifier values represent the ratio of each buyer type’s historical average spend relative to a fan baseline, they are computed from historical buyer spend data store dint he real market database (table 1). Table 3 includes “segment” column showing the buyer modifier type (“brand”) alongside the deal price for each completed deal, confirming the buyer modifier is determined from historical deal data containing buyer spend amounts).
As per claim 8, Lawrence discloses wherein the one or more processors are configured to: determine the adjusted price per follower based on the determined price per follower and the determined buyer type modifier ([0040] In an optional step 210, if social media follower count is known, the system 100 may determine an adjusted PPF. For example, the one or more processors 104 of the platform server 102 may be configured to determine an adjusted PPF, using the valuation model 108, based on Equation 2 (Eqn. 2), which is shown and described below: Adjusted PPF=PPF×Buyer Modifier).
As per claim 9, Lawrence discloses wherein the buyer type modifier includes at least one of: a donor modifier, a sponsor modifier, a brand modifier, a fan modifier, or a collective modifier ([0041] The buyer modifier may include a donor modifier, sponsor modifier, brand modifier, fan modifier, a collective modifier).
As per claim 13, Lawrence discloses wherein the user identifier data includes at least one of:
a student athlete identifier, a professional athlete identifier, a retired athlete identifier, an agent identifier, or a coach identifier (paragraph 34, “identifier (e.g., student athlete, professional athlete, retired athlete, agent, coach, and the like), “), wherein the activity type data includes at least one of: a social media channel activity type, a digital media activity type, a graphical element activity type, or an in-person activity type (paragraph 34, “activity type (e.g., Twitter post, Twitter fleet, Facebook post, Facebook story, Facebook live, TikTok, Instagram Post, Instagram story, Instagram IGTV, Instagram reel, Youtube, Photo/video/audio creation, Podcast appearance, digital press interview, appearance/meet-and-greet, autograph signing, in-person interview, keynote speech, production shoot, sport demonstration, and the like),”).
As per claim 14, Lawrence discloses wherein the user channel identifier data includes at least one of: a social media channel handle or a social media channel profile link (paragraph 34, “social media handle/profile link to determine a current follower count (e.g., for a specified platform or across all known platforms), and the like.”, paragraph 36, “and a social media handle/profile link field. “).
As per claim 15, Lawrence discloses generate one or more Application Programming Interface requests for one or more social media platform servers based on the received social media channel handle or the received social media channel profile link; and retrieve the real-time current follower count for the user based on the generated one or more Application Programming Interface requests (paragraph 36, “ In this regard, such data may be determined by a communication between the server 102 and a social media platform (e.g., by an Application Programming Interface (API) request).”, paragraph 39, “For example, the user may input their social media handle or profile link such that the one or more processors 104 of the platform server 102 may be able to retrieve the user's real-time follower count.”).
As per claim 16, Lawrence discloses wherein the filter, using the valuation model, the received real market data based on the received user input data comprises: filtering, using the valuation model, the received real market data based on the user identifier data and the activity type data, wherein the user identifier data includes a student athlete identifier and the activity type data includes a social media channel activity type (paragraph 37, “In this example, the one or more processors 104 of the platform server 102 may be configured to filter the received real market data based on the student athlete identifier and social post activity type. In this regard, the calculated activity pricing (calculated in step 220) may provide an accurate estimate of a user's market value for a specific social post activity type based on relevant real market data corresponding to the student athlete market.”).
As per claim 17, Lawrence discloses generate one or more control signals configured to cause the display of the user device to display the determined suggested activity price ([0050] FIG. 4 illustrates a graphical user interface (GUI) 500 of the system 100, in accordance with one or more embodiments of the present disclosure. In embodiments, the user device 112 may display the calculated suggested activity price (from step 220) on display 114 via the GUI 400. For example, the GUI 400 may list a market range for each specific activity type (e.g., Facebook Live, Facebook Story, Instagram IGTV, Instagram Reel, Media Creation, Photo/video/audio creation, and the like), which is tailored for that specific user (e.g., based on the real market data and user input data)).
As per claim 18, Lawrence discloses wherein the user input data further includes sport data, the sport data including at least one of:
sport type data, institution data, league data, or division data (paragraph 34, sport type (e.g., football, women's basketball, men's basketball, and the like), institution (e.g., school name, team name, and the like), conference (e.g., Big 12, Big 10, and the like), league/division, social media handle/profile link to determine a current follower count (e.g., for a specified platform or across all known platforms), and the like.).
As per claim 19, Lawrence discloses wherein the one or more predetermined thresholds include at least one of: similar athlete, similar sport and institution, similar sport and conference, similar sport and league/division, similar institution, similar conference, or similar league/division (paragraph 44, “based on one or more predetermined thresholds (as shown by Table 2). The one or more predetermined thresholds may include, but are not limited to, similar athlete, sport and institution, sport and conference, sport and league/division, institution, conference, league/division, and the like. In this regard, the match table may include the closest matching activity based on the one or more predetermined thresholds such that the activity price determined in step 220 reflects the real market data.”, table 2).
Claim(s) 3 is/are rejected under 35 U.S.C. 103 as being unpatentable over Lawrence (US 20230016916) in view of Walters (US 20210264199) and Sullivan (US 20190012746), as disclosed in the rejection of claim 1, in further view of Beckerman (US 20140358694).
As per claim 3, Lawrence does not disclose but Walters discloses hyperparameters as tunable configuration aspects of a machine learning model and an automated, iterative hyperparameter tuning using bathes of candidate hyperparameter sets, training/testing the model for each set, evaluating a success metric. Although Walters does not use “log denominator” as a hyperparameter, it treats hyperparameters generically as any tunable configuration controlling the model’s behavior which explicitly provides a machine learning based framework for automatically tuning such hyperparameters.
However, although Walters does not disclose the log denominator, Beckerman discloses applying a logarithmic transformation to the follower count as part of the pricing feature engineering (paragraph 36-37, the total fans value is log transformed before feeding the models that output the price per action. Because a logarithm grows more slowly than the underlying linear variable, applying a log transformation to follower count necessarily decreases the rate at which the model price output grows relative to follower count.).
Therefore, it would have been obvious to one of ordinary skill in the art before filing date of the current invention to include the limitations above as taught by Beckerman in the teaching of Lawrence in view of Walters, since the claimed invention is merely a combination of old elements, and in the combination each element merely would have performed the same function as it did separately, and one of ordinary skill in the art would have recognized that the results of the combination were predictable.
Claim(s) 10-12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Lawrence (US 20230016916) in view of Walters (US 20210264199) and Sullivan (US 20190012746), as disclosed in the rejection of claim 1, in further view of Lei (US 20190188536).
As per claim 10, Lawrence discloses partitioning the dataset into functional subsets including a reserved market influence portion (paragraph 47). However, Lawrence does not disclose but Lei discloses split the adjusted dataset into a training subset and a testing subset based on a predetermined split ratio, wherein the training subset is used for training the trained machine learning classifier of the valuation model for parameter adjustment of the one or more dynamic parameters, wherein the testing subset is used for validation of parameter performance of the determined suggested activity price for the user ([0047] At 206, all or a subset of the sales history data for the product is extracted (e.g., two years of sales history for yogurt in the Baltimore, area). A randomly selected portion of the sales history data is used as a training dataset, and the remainder is used as a validation dataset. In one embodiment, 80% of the sales history is randomly selected to form the training dataset, and the remaining 20% is used to form the validation dataset.).
Therefore, it would have been obvious to one of ordinary skill in the art at the time of the invention to include the limitation above as taught by Lei in the teaching of Lawrence, in order to train an algorithm using the training dataset and the feature test set to generate a trained algorithm and calculate an early stopping metric using the trained algorithm and the validation dataset (Lei, abstract).
As per claim 11, Lawrence in view of Walters, Sullivan and Lei disclose all the limitation of claim 10. Lawrence does not disclose but Lei further discloses wherein the predetermined split ratio 80/20 train/test, wherein 80% of the adjusted dataset is the training subset and 20% of the adjusted dataset is the testing subset (([0047] At 206, all or a subset of the sales history data for the product is extracted (e.g., two years of sales history for yogurt in the Baltimore, area). A randomly selected portion of the sales history data is used as a training dataset, and the remainder is used as a validation dataset. In one embodiment, 80% of the sales history is randomly selected to form the training dataset, and the remaining 20% is used to form the validation dataset... [0084] In round A (row 601) all available data points are used for purposes of comparison with the inventive determinations. For rounds 1-5 (rows 602-606), sampling data is used to do the estimation (per 504 of FIG. 5) and the remaining testing data is used to test/validate the model (per 508 of FIG. 5). In one embodiment, the sampling data is 80% of the data points, and the testing data is the remaining 20% of the data.)(please see claim 1 rejection for combination rationale).
As per claim 12, Lawrence does not disclose but Lei discloses calculate a root mean square error metric for each parameter combination used when performing the automated parameter tuning ([0048] At 208, an early stopping metric that has been received/selected determines when the iterative process will be completed and the optimized feature set is determined. In one embodiment, mean absolute percentage error (“MAPF”) is used and is compared to an error threshold “e”. Further, a maximum number of iterations is received and also determines when the iterative process is completed if an optimized feature set is not determined. At 208, it is determined if the early stopping metric or the maximum number of iterations has been reached… [0052] At 216, the early stopping metric is calculated by applying the trained algorithm (i.e., the model) on the validation dataset and functionality continues at 208 where it is determined if the early stopping metric is below the threshold e or maximum iterations are reached. Therefore, steps 210, 212, 214 and 216 are repeated.” Under BRI, an “early stopping metric” calculated per parameter combination on the validation dataset is a performance error metric equivalent to RMSE);
storing the calculated root mean square error metric for each parameter combination in the memory ([0051] At 214, an algorithm is trained using the training data set from 206 and using the features of feature test set S (i.e., both the mandatory and optional features) to generate a trained algorithm (i.e., the model). The algorithm to be trained can be any desired algorithm such as disclosed above (e.g., linear regression, ANN, etc.)…[0052] At 216, the early stopping metric is calculated by applying the trained algorithm (i.e., the model) on the validation dataset and functionality continues at 208 where it is determined if the early stopping metric is below the threshold e or maximum iterations are reached. Therefore, steps 210, 212, 214 and 216 are repeated.” Storage of combination metrics in memory is a necessary and inherent step in an iterative selection process); and
selecting the parameters combination with a lowest root mean square error metric for subsequent suggested activity pricing determinations ([0048] At 208, an early stopping metric that has been received/selected determines when the iterative process will be completed and the optimized feature set is determined. In one embodiment, mean absolute percentage error (“MAPF”) is used and is compared to an error threshold “e”. Further, a maximum number of iterations is received and also determines when the iterative process is completed if an optimized feature set is not determined. At 208, it is determined if the early stopping metric or the maximum number of iterations has been reached… [0060] If the early stopping metric is reached at 208, then at 220 the optimized feature set is feature test set S of 212.)(please see claim 1 rejection for combination rationale).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to OMAR ZEROUAL whose telephone number is (571)272-7255. The examiner can normally be reached Flex schedule.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Kevin Flynn can be reached at 5712703108. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
OMAR . ZEROUAL
Examiner
Art Unit 3628
/OMAR ZEROUAL/Primary Examiner, Art Unit 3629
Prosecution Insights