DETAILED ACTION
A. This action is in response to the following communications: Transmittal of New Application filed 10/24/2024.
B. Claims 1-20 remains pending.
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 9-15 are rejected under 35 U.S.C. 101 because the claimed invention is directed to certain methods of organizing human activity without significantly more. The claim(s) 9 recite(s) “receiving, by at least one processor, data indicating a sequence of actions performed by a computer during a computing process; generating, predicting and configuring”, grouping of abstract ideas. The mere nominal recitation of a generic processor and generic off the shelf trained machine learning model does not take the claim out of the “certain methods of organizing human activity” grouping. Thus, the claim recites an abstract idea.
This judicial exception is not integrated into a practical application because the claim as a whole merely describes how to generally “apply” the concept of receive, generate and updating action information in a computer environment. The claimed “processor” and “model” are recited at a high level of generality and are merely invoked as tools perform an existing action update process. Even considered in combination, simply implementing the abstract idea on a generic computer with storage devices recited at a high level of generality is not a practical application of the abstract idea.The limitation viewing insights (data selections) on a generic user interface executed by a generic computer, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting “by a processor,” nothing in the claim element precludes the step from practically being performed in the mind. For example, but for the “by a processor” language, “in response to receiving” in the context of this claim encompasses the user manually organizing data.
The claim does not include additional elements that are sufficient to amount to the significantly more than the judicial exception because as noted previously, the processor and the computer in combination merely describe how to generally “apply” the concept of updating in a computer environment. The same applies here. (MPEP 2106.05(d). Thus, even when viewed as a whole, nothing in the claims adds significantly more (i.e. an inventive concept) to the abstract idea.
The claim is ineligible.
Claims 10-15 do not include elements that amount to significantly more than the abstract idea and are also rejected under the same rational.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1-2,4-10 and 12-20 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Dicken, Luke et al. (US Pub. 2022/0176252 A1), herein referred to as “Dicken”.
As for claims 1, 9 and 15, Dicken teaches. A method and corresponding other method of claim 9 and system of claim 15 comprising:
at least one processor; and
at least one non-transitory computer-readable medium storing instructions that, when executed by the at least one processor, cause the at least one processor to perform processing comprising (par. 255 is the hardware environment discussed that executes the software given):
Overview of Dickens: (par. 37) Some aspects of this disclosure provides for a game distribution and updating mechanism in which a compiled game engine is persistently installed on a client device, while game configuration values (which are variable to customize gameplay for different players or different identified groups of players) are provided to the client device in a separate standalone data download. In some examples, the game configuration values are provided to the client device as non-compiled data that defines game rules, the appearance and/or behavior of interactive in-game objects (also referred to herein as interactive content items), and configuration of respective game levels (for example including a gameboard layout, performance targets, and experience points and/or coin rewards for respective performance targets).
receiving, by at least one processor, user data indicating a sequence of user interface (UI) actions (par. 61 summarizes the functionality of how the system can aid the user using the software user interface which in this case is a game; FIG. 5 is a data flow diagram 500 for game configuration value generation, in accordance with some examples. As shown, a player model 502 and content item(s) 504 are ingested by a game assessment engine 506. The game assessment engine 506 uses the player model 502 to generate a user action sequence 508, forecasting actions that a user, represented by the player model 502, would take in response to the content item(s) 504 being presented in a computer-implemented game. The game server 112 determines outcome(s) 510, specified in the software of the computer-implemented game, of the generated user action sequence 508. The user action sequence 508 and the outcome(s) 510 are ingested by the optimization engine 512. The optimization engine 512 optimizes configuration value(s) 514 for the content item(s) 504. The optimized configuration values 514 are stored in a database 516 for consumption by client device(s) 102 associated with the player model 502 (e.g., based on the user account(s) of the client device(s)).;
generating, by the at least one processor, a unified embedding from the sequence of UI actions, the unified embedding encoding: respective embeddings of the respective UI actions encapsulating UI action characteristics and sequential positions (par. 206-207 encoding by converting data (e.g. labels) to numerical data wherein each label is converted to a vector and the vectors is referred to as “One-Hot Embedding Matrix” for mapping converting a game action to vector with help of the encoding. To enable automated analysis of game traversal actions using a machine learning model, the action labels are to be converted to numerical values. Simple natural number representation is avoided for implying ordinal relationship in the actions. An existing method to convert categorical data (e.g., labels) to numerical data is “One-Hot Encoding”, where each label is converted to a vector containing a list of binary values 0 and 1, each vector having a predefined number of values, with one of the values being 1 and the remainder of the values being 0. The ordinal position of each value is mapped to a respective predefined attribute or parameter, in this instance being mapped to a respective predefined game action. Suppose for example a predefined set of 1000 distinct game actions, a 1000×1000 diagonal matrix can be constructed to represent the whole action feature space, where each game action is represented by a vector (column) of length 1000 with values 0 and 1. Each vector will have exactly one 1 and the rest of the values to be 0. The 1000×1000 matrix in referred to as the “One-Hot Embedding Matrix”, which is used to provide a mapping for converting a game action to its corresponding representation vector.), and
respective importance scores of the respective UI actions (par. 218-223 the player model stores scores/ data related to player motivations to calculate probability of attention; par. 224 At operation 2110, each player journey in the training set is mapped, using the previously constructed action embedding matrix, to a corresponding sequence of action embedding vectors These aspects of the method are again illustrated using the example player journey 2302 and the example action embedding matrix 2306. As shown schematically in FIG. 23C, each of the actions 2304 in player journey 2302 is, via action embedding matrix 2306, converted to a respective representation vector or action embedding vector 2310. For example, action [Collected Rewards] will be mapped to the vector ecollect with value [0.98, 0.21, . . . , 0.09]T, and action [Won a Level] will be mapped to the vector ewin with value [0.12, 0.42, . . . , 0.34]T, etc.).;
predicting, by the at least one processor processing the unified embedding with a trained machine learning (ML) model, a next likely action by the user within the UI (Par. 208 taking the “One Hot Embedding Matrix” and converting it to a significantly denser and low-dimensional vector for each game action by use of a neural network and further is referred to as action embedding; par. 210-214 give example technique for deriving action embedding matrix using player journey data to make predictions for what next action in the journey is (e.g. Used Hints Powerup);
and
Specifically for claim 1 modifying, by the at least one processor, the UI to add or change a UI element that enables the next likely action in response to the predicting and specifically for claim 9 configuring, by the at least one processor, the computer to perform the next likely action in response to the predicting (par. 218 the user interface is modified after the RNN training, this paragraph highlights the goal which is for automated content generation based upon parametric player model; FIG. 21 further comprises, operation 2124, retrieving a subset of player journeys from the graph database 2128 for use as a training set for an example and, at operation 2126, assigning respective labels with respect to the relevant psychological feature to respective items in the player journeys. In instances where such psychological features are to be estimated for input journeys (e.g., to estimate a player motivations for a journey, as in the example of FIG. 24B) labels are mapped to journeys. The Many-to-One player journey described below with reference to FIG. 24B is most useful when attempting to generate a single score/label to summarize a given journey. Such a many-to-one model serves to analyze the player's motivation to play the game. In this scenario, the label being produced in this model (and assigned to each player journey 2302 in the training set, at operation 2126) is in this example one the following four motivating factors, which respectively inform custom configuration of game levels as discussed previously with respect to automated content generation based on a parametric player model).
.
Difference in Claim 9: removes the mention of actions being related to user interface actions and states they are within a computing process instead the analysis of user interface actions covers this broader limitation.
As for claim 2, 10 and 16, Dickens teaches. The method of claim 1, wherein the generating comprises generating the respective embeddings by performing, for each of the respective UI actions, processing comprising:
transforming the respective UI action into an initial embedding vector representing at least one feature of the respective UI action;
generating a positional encoding of the respective UI action; and
combining the initial embedding vector and the positional encoding to form the respective embedding (par. 218-225 At operation 2112 (FIG. 21), the action embedding vectors are used to train a recurrent neural network model (in this example, RNN 2402 in FIG. 24A and FIG. 24B) by feeding the action embedding vectors 2310 sequentially based on their original temporal order in the player journey 2302, the RNN 2402 being trained to conform such inputs to the corresponding assigned motivation/emotional state label (also referred to for short as a psych label). This RNN model allow the production of a single player label of interest or a temporal sequence of labels associated with that player throughout the given journey, depending on the choice of a many-to-many structure (FIG. 24A) or a many-to-one structure (FIG. 24B)).
As for claims 4, 12 and 17, Dicken teaches. The method of claim 1, wherein the generating comprises generating the respective importance scores by performing, for each of the respective UI actions, processing comprising: determining respective similarity scores of respective pairs of the respective embeddings; and calculating a probability distribution containing the respective importance scores using an attention mechanism taking the respective similarity scores as inputs (par. 218-223 the player model stores scores/ data related to player motivations to calculate probability of attention; par. 224 At operation 2110, each player journey in the training set is mapped, using the previously constructed action embedding matrix, to a corresponding sequence of action embedding vectors These aspects of the method are again illustrated using the example player journey 2302 and the example action embedding matrix 2306. As shown schematically in FIG. 23C, each of the actions 2304 in player journey 2302 is, via action embedding matrix 2306, converted to a respective representation vector or action embedding vector 2310. For example, action [Collected Rewards] will be mapped to the vector ecollect with value [0.98, 0.21, . . . , 0.09]T, and action [Won a Level] will be mapped to the vector ewin with value [0.12, 0.42, . . . , 0.34]T, etc.).
As for claims 5 and 13, Dicken teaches. The method of claim 4, wherein the attention mechanism includes a softmax function (softmax is not specifically mentioned but through same function that is disclosed it is inherent the prior art is discussing a softmax function, a common term in mathematics as known in the art. Definition of the softmax function is a mathematical operation commonly used in neural networks for multi-class classification tasks. It transforms a vector of raw scores (logits) into a probability distribution, where the sum of all probabilities equals 1; the prior art teaches in par. 214 The candidate action embedding vectors for respective preceding actions 2304 of the player journey 2302, derived from the candidate embedding matrix E for the identified actions 2304, are then fed as training input to a Neural Net Model (see example NN Model 2308 in FIG. 23B) The mapping matrix E is the parameter that that is thus being fit to the NN Model 2308 to maximize the probability of predicting the correct the next action to be [Used Hints Powerup] and Paragraph 227 The RNN model training (at operation 2112) in the many-to-many structure of FIG. 24A in this example proceeds as follow:
Step 1: Take the first action in the player journey [Collected Rewards] and feed its embedding vector to construct the first hidden layer of RNN 2402. Output a predicted label (Label 1) using the information in Layer 1. The first output from RNN 2402 for each layer is a predicted probability distribution. e.g., [0.1, 0.6; 0.3], where each number represent the probability of an emotional state (e.g., [Happy, Sad, Frustrated], and the sum of these three numbers is 1 (equal to 100% probability). It is inherent that the prior art is referencing softmax function.
As for claims 6, 14 and 18, Dicken teaches. The method of claim 1, wherein:
the ML model comprises a recurrent neural network (RNN); and
the predicting comprises modeling the sequence by the RNN using the unified embedding as input and predicting at least one hidden state from the modeling, the at least one hidden state indicating the next likely action (par. 227-232 RNN model training modeling sequence using embeddings as input (feeding) to construct first hidden layer of RNN 2402).
As for claims 7 and 19, Dicken teaches. The method of claim 1, wherein the modifying comprises fetching at least one component or parameter of the UI element from at least one data store and configuring the UI to include the at least one component or parameter (par. 227-232 the user interface is modified based upon these steps in the RNN model training and based upon the user input (playing the game), wherein Step 2 discusses Use the action embedding vector 2310 for the second action 2304 [Won a Level] together with compacted information [a1] passed from Layer 1 to construct a second hidden layer in the neural network. Output a predicted label from Layer 2. Unlike Layer 1, Layer 2 is constructed using inputs from two sources: compacted information from Layer 1; and action vector 2310 from the player journey 2302. Step 3: Repeat Step 2 for each but the last action, continuing the same procedure to construct a new hidden neural network layer at each action.
Last Step: Use the embedding vector 2310 for the last action 2304 [Used Hints Powerup] together with the information passed from the previous layers to construct the last hidden layer in the neural network, thus producing a predicted label from the last layer).
As for claims 8 and 20, Dicken teaches. The method of claim 7, wherein the modifying further comprises configuring a variable portion of the UI element with context data determined from the user data (par. 218 the user interface is modified after the RNN training, this paragraph highlights the goal which is for automated content generation based upon parametric player model; FIG. 21 further comprises, operation 2124, retrieving a subset of player journeys from the graph database 2128 for use as a training set for an example and, at operation 2126, assigning respective labels with respect to the relevant psychological feature to respective items in the player journeys. In instances where such psychological features are to be estimated for input journeys (e.g., to estimate a player motivations for a journey, as in the example of FIG. 24B) labels are mapped to journeys. The Many-to-One player journey described below with reference to FIG. 24B is most useful when attempting to generate a single score/label to summarize a given journey. Such a many-to-one model serves to analyze the player's motivation to play the game. In this scenario, the label being produced in this model (and assigned to each player journey 2302 in the training set, at operation 2126) is in this example one the following four motivating factors, which respectively inform custom configuration of game levels as discussed previously with respect to automated content generation based on a parametric player model).
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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) 3 and 11is/are rejected under 35 U.S.C. 103 as being unpatentable over Dickens in view of Vaidya, Suthirth et al. (US Pat. 12,266,107 B1), herein referred to as “Vaidya”
As for claim 3 and 11, Dickens teaches. The method of claim 2. But does not specifically teach wherein generating the positional encoding is performed using at least one of a sine function and a cosine function; however in the same field of endeavor Yaidya teaches (positional encoding may include mapping each position in the sequence to a position vector. In some embodiments, trigonometric functions, such as sine and cosine, may be used to determine the values in the position vector. In some embodiments, position vectors for a plurality of positions in a sequence may be assembled into a position matrix, wherein each row of position matrix may represent a position in the sequence at col. 18, lines 10-15).
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine Vaidya into Dickens because Vaidya suggests in some embodiments, long short-term memory (LSTM) and related recurrent neural networks (RNNs) and Time Delay Neural Networks (TDNN's) may be used for automatic speech recognition, for example over longer time intervals for continuous speech recognition (col. 8, line 21) and traditional methods rely heavily on manual processes and the expertise of the physician. This can lead to variability in outcomes (col. 1, line 16).
(Note:) It is noted that any citation to specific, pages, columns, lines, or figures in the prior art references and any interpretation of the references should not be considered to be limiting in any way. A reference is relevant for all it contains and may be relied upon for all that it would have reasonably suggested to one having ordinary skill in the art. In re Heck, 699 F.2d 1331, 1332-33, 216 USPQ 1038, 1039 (Fed. Cir. 1983) (quoting In re Lemelson, 397 F.2d 1006,1009, 158 USPQ 275, 277 (CCPA 1968)).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
GUI PROCESSING AND GENERATION SYSTEMS FOR VISUAL USER INTERFACE DEPICTION AND IMPROVEMENT EXPLANATION
Document ID
US 20250390905 A1
Date Published
2025-12-25
Abstract
Systems and methods access stored object information of a tangible object associated with a user account of a user, the stored object information being stored to data storage location(s) and indicating parameters associated with use of the tangible object. Parameter(s) are ascertained from the stored object information, with at least one parameter including benefit(s) available to the user upon use of the tangible object. Customized interface content to be depicted via a GUI of the user device is generated, where the customized interface content represents a virtual depiction of the tangible object and at least one of the benefit(s) available upon use of the tangible object. Display of the customized interface content via the GUI of the user device is initiated, where the customized interface content depicts at least one of the benefit(s) as part of a design of the virtual depiction of the tangible object.
Predictive Learner Recommendation Platform
Document ID
US 11928607 B2
Date Published
2024-03-12
Abstract
Embodiments are directed to managing data for a predictive learner recommendation platform. A platform that includes applications hosted in an application layer may be provided. The applications may be employed to provide a request to determine a pathway prediction for a learner such that the pathway prediction may be associated with a role offered by employers. Prediction engines associated with the request may be determined based on the service layer interface and the request such that the request may be provided to the determined prediction engines via the service layer interface. The prediction engines may be employed to generate the pathway prediction based on a learner profile that corresponds with the learner, a role success profile that corresponds to the employers, and models that are trained to predict matches between the learner profile and the role success profile.
Inquires
Any inquiry concerning this communication should be directed to NICHOLAS AUGUSTINE at telephone number (571)270-1056.
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.
PNG
media_image1.png
213
559
media_image1.png
Greyscale
/NICHOLAS AUGUSTINE/Primary Examiner, Art Unit 2178 July 31, 2026
Prosecution Insights