Detailed Action
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Notice to Applicant
The following is a Final Office action to Application Serial Number 18/780,999, filed on July 23, 2024. In response to Examiner’s Non-Final Office Action of November 6, 2025, Applicant, on February 6, 2026, amended claims 1, 9, 11 and 19-20; and cancelled claims 2, 5, 12, and 15; added claims 21-24. Claims 1, 3-4, 6-11, 13-14, 16-24 are pending in this application and have been rejected below.
Response to Amendment
Applicant’s amendments are acknowledged.
Regarding 35 U.S.C. § 101 rejection, the amendment has been considered and is insufficient to overcome the rejection.
The 35 U.S.C. § 103 rejections are hereby amended pursuant to applicants
amendments. Updated 35 U.S.C. § 103 rejections have been applied to amended
claims. Please refer to the § 103 rejection for further explanation and rationale.
Response to Arguments
Applicant’s arguments filed February 6, 2026 have been fully considered but they are not persuasive and/or are moot in view of the revised rejections. Applicant’s arguments will be addressed herein below in the order in which they appear in the response filed February 6, 2026.
On page 10-11 of the Remarks regarding 35 U.S.C. § 101, Applicant states the amended claims recite specific technological steps that cannot be performed mentally. In response, regarding the 35 U.S.C. § 101 rejection, Examiner finds under the broadest reasonable interpretation receiving delivering information; and performing discrepancy analysis falls within the Abstract idea grouping of “Mental Processes” – evaluation. The claims primarily recite the additional element of using computer components to perform each step. The “computer readable medium”, “memory”, “processor”, and “computer system” is recited at a high-level of generality, such that it amounts no more than mere instructions to apply the exception using a computer component. See MPEP 2106.05(f). The general use of a machine learning analysis does not provide a meaningful limitation to transform the abstract idea into a practical application.
On page 11-13 of the Remarks regarding 35 U.S.C. § 101, Applicant states amended claims integrate any alleged abstract idea into a practical application by solving the specific technological problem of computational resource optimization in delivery systems. Further, the claims include additional elements sufficient to amount to significantly more than the judicial exception and that the additional elements of computer readable medium, memory, processor, and computer system are insufficient to amount to significantly more. The claims require a machine learning model trained for delivery discrepancy analysis, automated feature extraction from physical objects, predictive analysis of user engagement based on discrepancy characteristics, and real-time decision-making based on model outputs. In response, regarding the 35 U.S.C. § 101 rejection, the claims primarily recite the additional element of using computer components to perform each step. The “computer readable medium”, “memory”, “processor”, and “computer system” is recited at a high-level of generality, such that it amounts no more than mere instructions to apply the exception using a computer component. See MPEP 2106.05(f). Furthermore, the claim 1, claim 11 and claim 20 recite using one or more machine learning analysis techniques. The specification discloses the machine learning analysis at a high-level of generality, providing examples of different techniques that may be applied. The general use of a machine learning analysis does not provide a meaningful limitation to transform the abstract idea into a practical application. Therefore, currently, the machine learning is solely used a tool to perform the instructions of the abstract idea.
On Pg. 13-14 regarding the 35 U.S.C. § 103 rejection, Applicant argues the latent space embedding and similarity metric limitations of Claims 5 and 15 are not taught or suggested by the cited references, either individually or in combination. In response, Examiner disagrees. The limitations of claim 5 and claim 15 are recited very broadly without specific algorithmic support. Brown discloses conversion Par. 23- Vector space methods provide an encoding scheme to convert information into points in a high-dimensional space where nearness in the space reflects factors such as relevance, relatedness, or semantic similarity. More generally, these vector spaces can form an embedding which is a mathematical structure that imposes a relationship among objects. Par. 24 discloses “From a machine learning perspective, these methods can utilize supervised learning techniques in that “side-information” of similarity or neighbor labels are prescribed a priori. A learning metric process can be obtained that is unsupervised. To do this, the temporal correlations in visual sensory data can be leveraged by manipulating the data as a time-ordered sequence of images, where adjacent image pairs form positive training examples that exhibit both similarity of perceptual features and nearness of physical locality. Such a process can exploit temporal contiguity and constrain perceptual organization in both space and time. Such an organization might also facilitate an intuitive representation of temporal context.”; Par. 27 discloses “A goal of such a process can be to capture sensory relatedness such that similar features map to nearby points in a vector space embedding….The feature vectors, having a second size, can then be fed to an embedding network 406 as training data. Such a process of feature extraction followed by feature embedding produces a ‘relative space’ representation, or visualization 408, that is topologically consistent with the latent structure of the sensory information and draws a compelling analogy with the complementary nature of the visual cortex and the hippocampus.”
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, 3-4, 6-11, 13-14, 16-24 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Claims 1, 3-4, 6-11, 13-14, 16-24 are directed to determining whether to perform interventions by determining magnitude of the discrepancies.
Claim 1 recites a method for determining whether to perform interventions by determining magnitude of the discrepancies, Claim 11 recites an article of manufacturing for determining whether to perform interventions by determining magnitude of the discrepancies and Claim 20 recites a system for determining whether to perform interventions by determining magnitude of the discrepancies, which include receiving, based on user input into an application, a request for delivery of a first physical object; receiving information from a scanning system identifying a second physical object as satisfying the request for the first physical object; detecting a discrepancy based on detecting that the second physical object is not a same physical object as the first physical object; receiving, as output from the machine learning model, a measure of predicted future engagement of the user with the application based on the discrepancy; and instructing the application to output an intervention based on the measure of predicted future engagement of the user.
As drafted, this is, under its broadest reasonable interpretation, within the Abstract idea grouping of “Mental Processes” – evaluation. The recitation of “computer readable medium”, “memory”, “processor”, and “computer system”, provide nothing in the claim elements to preclude the step from being “Mental Processes”- evaluation. Accordingly, the claim recites an abstract idea.
This judicial exception is not integrated into a practical application. The claims primarily recite the additional element of using computer components to perform each step. The “computer readable medium”, “memory”, “processor”, and “computer system” is recited at a high-level of generality, such that it amounts no more than mere instructions to apply the exception using a computer component. See MPEP 2106.05(f). Furthermore, the claim 1, claim 11 and claim 20 recite using one or more machine learning analysis techniques. The specification discloses the machine learning analysis at a high-level of generality, providing examples of different techniques that may be applied. The general use of a machine learning analysis does not provide a meaningful limitation to transform the abstract idea into a practical application. Therefore, currently, the machine learning is solely used a tool to perform the instructions of the abstract idea. Accordingly, the additional elements do not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claims also fail to recite any improvements to another technology or technical field, improvements to the functioning of the computer itself, use of a particular machine, effecting a transformation or reduction of a particular article to a different state or thing, and/or an additional element applies or uses the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment, such that the claim as a whole is more than a drafting effort designed to monopolize the exception. See 84 Fed. Reg. 55. In particular, there is a lack of improvement to a computer or technical field in discrepancy analysis.
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the additional elements when considered both individually and as an ordered combination do not amount to significantly more than the abstract idea. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of “computer readable medium”, “memory”, “processor”, and “computer system” is insufficient to amount to significantly more. (See MPEP 2106.05(f) – Mere Instructions to Apply an Exception – “Thus, for example, claims that amount to nothing more than an instruction to apply the abstract idea using a generic computer do not render an abstract idea eligible.” Alice Corp., 134 S. Ct. at 235). Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept.
The claim fails to recite any improvements to another technology or technical field, improvements to the functioning of the computer itself, use of a particular machine, effecting a transformation or reduction of a particular article to a different state or thing, adding unconventional steps that confine the claim to a particular useful application, and/or meaningful limitations beyond generally linking the use of an abstract idea to a particular environment. See 84 Fed. Reg. 55. Viewed individually or as a whole, these additional claim element(s) do not provide meaningful limitation(s) to transform the abstract idea into a patent eligible application of the abstract idea such that the claim(s) amounts to significantly more than the abstract idea itself. With regards to receiving data and step 2B, it is M2106.05(d)- Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information) and Storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015).
Examiner concludes that the additional elements in combination fail to amount to significantly more than the abstract idea based on findings that each element merely performs the same function(s) in combination as each element performs separately. The claim is not patent eligible. Thus, taken alone, the additional elements do not amount to significantly more than the above-identified judicial exception (the abstract idea). Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually.
Dependent Claims 3-4, 6-10, 13-14, 16-19 and 21- 24 inputting the features of the first physical object and the features of the second physical object into the machine learning model occurs responsive to detecting the discrepancy; inputting the features of the first physical object and the features of the second physical object into the machine learning model occurs responsive to receiving further user input from the user regarding the second physical object; instructing the application to output the intervention occurs responsive to that the measure of predicted future engagement of the user falls below a threshold; the intervention comprises the application outputting, on a user interface of a client device, an indication of a remedial action; the application is configured to not output the intervention responsive to that the measure of predicted future engagement of the user falls above a threshold; training the machine learning model by: accessing a training dataset comprising historical information on past discrepancies, user characteristics, order histories, and user feedback data; preprocessing the training dataset; splitting the preprocessed training dataset into a training subset and a validation subset; training the machine learning model based on the training subset by: inputting the training subset into the machine learning model; receiving, from the machine learning model, predicted measures of future engagement of users with the application based on the inputting; comparing the predicted measures of future user engagement with actual engagement data in the validation subset; calculating a loss function based on the comparison; adjusting parameters of the machine learning model to minimize the loss function; and iterating the training process until a predetermined performance threshold is met; collecting user feedback data responsive to the application outputting the intervention, wherein the user feedback data comprises at least one of: user actions taken within the application following the intervention, and user engagement metrics with the application after the intervention; storing the collected user feedback data in a database; re-training the machine learning model based, at least in part, on the collected user feedback data, wherein retraining comprises at least: incorporating the collected user feedback data into the training dataset; and further narrowing the abstract idea. These recited limitations in the dependent claims do not amount to significantly more than the above-identified judicial exceptions in Claims 1, 11 and 20. Regarding Claims, 7 and 17, and the additional elements of “client device” and “user interface”- it is M2106.05(d)- Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information). Regarding claim 3-4, 9-10, 13-14,19 and 24, and the additional element of machine learning model - the specification discloses the machine learning at a high-level of generality, providing examples of different techniques that may be applied. The general use of a machine learning technique does not provide a meaningful limitation to transform the abstract idea into a practical application. Therefore, currently, the machine learning is solely used a tool to perform the instructions of the abstract idea.
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 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.
Claims 1, 3-4, 6-11, 13-14, 16-24 are rejected under 35 U.S.C. 103 as being unpatentable over Harris et al., US Patent No. 10922349B1, [hereinafter Harris], in view of Brown et al., US Publication No. 20210142491A1, [hereinafter Brown], and in view of Roy et al., US Publication No. 20210241289A1, [hereinafter Roy],
Regarding Claim 1,
Harris teaches
A computer-implemented method, comprising: receiving, based on user input into an application, a request for delivery of a first physical object (Harris Col 3 Ln 30-55 The access device 192 may present an interface including one or more control elements to receive delivery information status request inputs. The status request inputs may include one or more of item, delivery, or location information for which delivery status is requested.”) ;
receiving information from a scanning system identifying the second physical object as satisfying the request for the first physical object; (Harris Col 6Ln46-66-The delivery agent 120 may interact with one or more interfaces presented via the delivery device 122. The interfaces may include control elements to receive input or adjust a function of the delivery device 122. For example, the delivery device 122 may include a camera or other optical scanning element. The delivery device 122 may activate the camera to scan an item 124 to be delivered. For example, the item 124 may include a scannable code or other detectable indicator. The scannable code may encode or indicate an identifier for the item or delivery information related to the item such as address, unique identifier of a monitoring device 150, or other information used in identifying or delivering the item 124.; Col 8 Ln 5-25- For example, some physical locations may have ornamental features that are mistaken for boxes. In such instances, the delivery information system 200 may incorrectly associate the delivery event with monitoring data associated with a non-delivery monitoring event. The feedback can be included during automatic identification to select a vision model that is adapted to more accurately identify the portions of the monitoring data associated with delivery events. For example, a general vision model may be used which can quickly identify portions for a majority of the instances of monitoring data. However, for physical locations where negative feedback is received, a slower but more accurate model may be used to process the monitoring data generated by a monitoring device associated with the location.; Col 9-10- At block 510, the delivery information system 200 may determine whether the delivery event is associated with a monitoring threshold. The monitoring threshold may identify a size of the portion of monitoring data for presenting the delivery event. For example, for a delivery event it may be desirable to include two minutes of monitoring data whereas for a security event it may be desirable to show fifteen minutes or more of the monitoring data. The threshold may be defined using a delivery information configuration stored in memory. In some implementations, the threshold may be specific to one or more of a user, a physical location, or a monitoring device.”)
detecting a discrepancy based on detecting that the second physical object is not a same physical object as the first physical object (Harris col 7, In 40-62, The object detection may include selecting a computer vision model trained to detect delivery items from one or more images. The model may receive, as inputs, image data and, in some implementations, estimated dimensions of the delivered item. The model may provide as an output detection information indicating whether an object (e.g., a delivered item) is shown in the images provided to the model. The output information may include a detection result indicating a likelihood that an image shows an item associated with the delivery event; Col 8 Ln 5-2; col 9-10- The analysis result may be used at block 530 to determine whether the item is represented in the monitoring data. The determination may compare an output value from the model to a detection threshold. );
inputting a first set of features of the first physical object and a second set of features of the second physical object into a machine learning model (Harris col 7, In 40-62, The object detection may include selecting a computer vision model trained to detect delivery items from one or more images. The model may receive, as inputs, image data and, in some implementations, estimated dimensions of the delivered item. The model may provide as an output detection information indicating whether an object (e.g., a delivered item) is shown in the images provided to the model. The output information may include a detection result indicating a likelihood that an image shows an item associated with the delivery event - If the determination at block 530 is negative, the method 500 may proceed to block 535 to transmit the delivery event information for further review The review system may include sophisticated machine learning models (e.g., a vision model for image monitoring data- the failure may be due to an improper delivery (e.g., item left in an inappropriate location or unauthorized removal of the item from the physical location), col 9-10);
Harris fails to teach the following feature taught by Brown:
converting the features of the first physical object and the features of the second physical object into latent space embeddings; generating a similarity metric based on the latent space embeddings; and inputting the similarity metric into the machine learning model. (Brown Par. 23- “A computational analogue of these relative spaces involves vector space models. Vector models often find use in information retrieval and natural language processing as methods to encode information according to some basis decomposition that may be derived from keywords or phrases that capture semantic variance. Vector space methods provide an encoding scheme to convert information into points in a high-dimensional space where nearness in the space reflects factors such as relevance, relatedness, or semantic similarity. More generally, these vector spaces can form an embedding which is a mathematical structure that imposes a relationship among objects. When an embedding imposes a relationship between objects using distance it creates a metric space which induces a topology.; Par. 24-From a machine learning perspective, these methods can utilize supervised learning techniques in that “side-information” of similarity or neighbor labels are prescribed a priori. A learning metric process can be obtained that is unsupervised. To do this, the temporal correlations in visual sensory data can be leveraged by manipulating the data as a time-ordered sequence of images, where adjacent image pairs form positive training examples that exhibit both similarity of perceptual features and nearness of physical locality. Such a process can exploit temporal contiguity and constrain perceptual organization in both space and time. Such an organization might also facilitate an intuitive representation of temporal context. Par. 27 discloses “A goal of such a process can be to capture sensory relatedness such that similar features map to nearby points in a vector space embedding….The feature vectors, having a second size, can then be fed to an embedding network 406 as training data. Such a process of feature extraction followed by feature embedding produces a ‘relative space’ representation, or visualization 408, that is topologically consistent with the latent structure of the sensory information and draws a compelling analogy with the complementary nature of the visual cortex and the hippocampus.”; Par. 23-27”);
Harris and Brown are directed to model analysis. It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have improve upon data analysis of Harris, as taught by Brown, by utilizing additional data analysis with a reasonable expectation of success of arriving at the claimed invention. One of ordinary skill in the art would have been motivated to make the modification to the teachings of Harris with the motivation of improving precision ( Brown Par. 29).
Harris and Brown teach model analysis and the feature is expounded upon by Roy:
receiving, as output from the machine learning model, a measure of predicted future engagement of the user with the application based on the discrepancy (Roy Par. 19; Par. 37; Par. 53- “The process 200 continues by comparing the second set of user interaction data and the second set of user sentiment data to the satisfaction threshold. This is illustrated at step 220. In embodiments, the system may use a scoring model when comparing the data. For example, the second set of user interaction data and the second set of user sentiment data may be provided a score and that score will be compared to the scores used to generate the satisfaction threshold (as described in FIG. 3). In some embodiments, the first set of user interaction data and the first set of user sentiment data may be compared directly to the second set of user interaction data and the second set of user sentiment data, respectively (e.g., without generating a score). The process 200 continues by determining if the satisfaction threshold has been exceeded.”);
and instructing the application to output an intervention based on the measure of predicted future engagement of the user. (Roy Par. 17-“Embodiments of the present disclosure relate to a system for determining if a user (e.g., customer) is satisfied with a product and, if not, mitigating the reason(s) for dissatisfaction with the product before the customer relationship with the product's maker and/or service provider is strained.”; Par. 65-“The process 400 continues by implementing one of the set of historical actions when outputting the action (at step 235 of process 200) to reduce dissatisfaction of the user. This is illustrated at step 415. In this way, the system may continually predict the best strategies for obtaining a positive outcome on future occasions when outputting actions for reducing user dissatisfaction with the product.”)
Harris, Brown and Roy are directed to model analysis. It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have improve upon model analysis of Harris in view of Brown, as taught by Roy, by utilizing product interaction analysis with a reasonable expectation of success of arriving at the claimed invention. One of ordinary skill in the art would have been motivated to make the modification to the teachings of Harris in view of Brown with the motivation of improved customer relationship between the user and the product's maker, service provider, and/or brand ( Roy Par. 25).
Regarding Claim 2 and Claim 12- Cancelled
Regarding Claim 3 and Claim 13, Harris in view of Brown in further view of Roy teach The method of claim 1, …, and The non-transitory computer-readable medium of claim 11, …
wherein inputting the features of the first physical object and the features of the second physical object into the machine learning model occurs responsive to detecting the discrepancy (Harris Col 6 & 7- The feedback can be included during automatic identification to select a vision model that is adapted to more accurately identify the portions of the monitoring data associated with delivery events. For example, a general vision model may be used which can quickly identify portions for a majority of the instances of monitoring data. However, for physical locations where negative feedback is received, a slower but more accurate model may be used to process the monitoring data generated by a monitoring device associated with the location.”)
Regarding Claim 4 and Claim 14, Harris n view of Brown in further view of Roy teach The method of claim 1, …, and The non-transitory computer-readable medium of claim 11, …
wherein inputting the features of the first physical object and the features of the second physical object into the machine learning model occurs responsive to receiving further user input from the user regarding the second physical object. (Harris Col 8- Because the portion of monitoring data presented may be automatically identified by the delivery information system 200, the interface 400 may include a feedback control element 444. The feedback control element 444 may be implemented as a button, selector, or text input element to submit feedback on the identified portion of the monitoring data. The feedback may be used to refine the delivery identification for subsequent deliveries. For example, some physical locations may have ornamental features that are mistaken for boxes. In such instances, the delivery information system 200 may incorrectly associate the delivery event with monitoring data associated with a non-delivery monitoring event. The feedback can be included during automatic identification to select a vision model that is adapted to more accurately identify the portions of the monitoring data associated with delivery events. For example, a general vision model may be used which can quickly identify portions for a majority of the instances of monitoring data. However, for physical locations where negative feedback is received, a slower but more accurate model may be used to process the monitoring data generated by a monitoring device associated with the location.”)
Regarding Claim 5 and Claim 15- Cancelled
Regarding Claim 6 and Claim 16 and Claim 21, Harris in view of Brown in further view of Roy teach The method of claim 1, …, and The non-transitory computer-readable medium of claim 11, … and The computer system of claim 20…
Harris in view of Brown teach model analysis and the feature is expounded upon by Roy:
wherein instructing the application to output the intervention occurs responsive to that the measure of predicted future engagement of the user falls below a threshold. (Roy Par 3- A processor may collect a first set of user interaction data related to the product from a device on a network and a first set of user sentiment data related to the product from a communication channel. The user interaction data and the user sentiment data are specific to a user. The processor may generate a user profile for the user, the user profile including a satisfaction threshold for using the product based in part on the first set of user interaction data and the first set of sentiment data. The processor may monitor a second set of user interaction data related to the product from the device on the network and a second set of user sentiment data related to the product from the communication channel. The processor may compare the second set of user interaction data and the second set of user sentiment data to the satisfaction threshold. The processor may determine that the user is experiencing dissatisfaction with the product when the satisfaction threshold has been exceeded. In response to the satisfaction threshold being exceeded, the processor may output an action to reduce dissatisfaction of the user.”)
Harris, Brown and Roy are directed to model analysis. It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have improve upon data analysis of Harris in view of Brown, as taught by Roy, by utilizing product interaction analysis with a reasonable expectation of success of arriving at the claimed invention. One of ordinary skill in the art would have been motivated to make the modification to the teachings of Harris in view of Brown with the motivation of improved customer relationship between the user and the product's maker, service provider, and/or brand ( Roy Par. 25).
Regarding Claim 7 and Claim 17 and Claim 22, Harris in view of Brown in further view of Roy teach The method of claim 6, …, and The non-transitory computer-readable medium of claim 16, … and The computer system of claim 21,…
Harris teaches product analysis and the feature is expounded upon by Roy:
wherein the intervention comprises the application outputting, on a user interface of a client device, an indication of a remedial action (Roy Par 34- Further, the social identifiers included in the user profile allow customer analysis device 102 to automatically output various actions (e.g., set up maintenance service for the product, sending discount coupons, offering guidance on how to user product, etc.) to the user when attempting to mitigate dissatisfaction with the given product. For example, customer analysis device 102 may send various actions directly to the user's smart phone, email, or user interface of an IoT device 120.”)
Harris, Brown and Roy are directed to model analysis. It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have improve upon data analysis of Harris in view of Brown, as taught by Roy, by utilizing product interaction analysis with a reasonable expectation of success of arriving at the claimed invention. One of ordinary skill in the art would have been motivated to make the modification to the teachings of Harris in view of Brown with the motivation of improved customer relationship between the user and the product's maker, service provider, and/or brand ( Roy Par. 25).
Regarding Claim 8 and Claim 18, Harris in view of Brown in further view of Roy teach The method of claim 1, …, and The non-transitory computer-readable medium of claim 11, … and The computer system of claim 21,…
Harris in view of Brown teach model analysis and the feature is expounded upon by Roy:
wherein the application is configured to not output the intervention responsive to that the measure of predicted future engagement of the user falls above a threshold. (Roy Fig. 2; Par. 47-54- Referring now to FIG. 2, shown is a flow diagram of an example process 200 for mitigating user dissatisfaction with a product, in accordance with embodiments of the present disclosure. The process 200 may be performed by processing logic that comprises hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, etc.), software (e.g., instructions run on a processor), firmware, or a combination thereof. In some embodiments, the process 200 is a computer-implemented process. The process 200 may be performed by processor 104 exemplified in FIG. 1.”)
Harris, Brown and Roy are directed to model analysis. It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have improve upon data analysis of Harris in view of Brown, as taught by Roy, by utilizing product interaction analysis with a reasonable expectation of success of arriving at the claimed invention. One of ordinary skill in the art would have been motivated to make the modification to the teachings of Harris in view of Brown with the motivation of improved customer relationship between the user and the product's maker, service provider, and/or brand ( Roy Par. 25).
Regarding Claim 9 and Claim 19 and Claim 24, Harris in view of Brown in further view of Roy teach The method of claim 1, …, and The non-transitory computer-readable medium of claim 11, … and The computer system of claim 21, …
Harris fails to teach the following feature taught by Brown:
further comprising training the machine learning model by: accessing a training dataset… (Brown Par. 16- The embedding network is trained in part by analyzing the triplet loss values. Once a trained model is obtained, an image of a current position of a vehicle, as well as image data for a target destination, can be provided as input to the trained embedding model, ..; Par. 24; Par 27)
preprocessing the training dataset (Brown Par. 69-“ In some embodiments, hyperparameters can be tuned in certain categories, as may include data preprocessing (in other words, translating words to vectors), CNN architecture definition (for example, filter sizes, number of filters), stochastic gradient descent parameters (for example, learning rate), and regularization (for example, dropout probability), among other such options.”); splitting the preprocessed training dataset into a training subset and a validation subset (Brown Par. 59; Par. 61-“ Shuffling can be performed in some embodiments before the training data set is split into training and evaluation subsets, such that a relatively even distribution of data types is utilized for both stages. In some embodiments the training manager can automatically shuffle the data using, for example, a pseudo-random shuffling technique.”); … calculating a loss function based on the comparison (Brown Par. 72); adjusting parameters of the machine learning model to minimize the loss function (Brown Par. 72); and iterating the training process until a predetermined performance threshold is met.(Brown Par. 60- “The model is evaluated to determine whether the model will provide at least a minimum acceptable or threshold level of performance in predicting the target on new and future data. Since future data instances will often have unknown target values, it can be desirable to check an accuracy metric of the machine learning on data for which the target answer is known, and use this assessment as a proxy for predictive accuracy on future data.”);
Harris and Brown are directed to model analysis. Brown improves upon the model analysis. It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have improve upon data analysis of Harris, as taught by Brown, by utilizing additional data analysis with a reasonable expectation of success of arriving at the claimed invention. One of ordinary skill in the art would have been motivated to make the modification to the teachings of Harris with the motivation of improving precision ( Brown Par. 29).
Harris in view of Brown fail to teach the following feature taught by Roy:
… comprising historical information on past discrepancies, user characteristics, order histories, and user feedback data (Roy Par. 63-“The process 400 begins by analyzing, using machine learning, historical user interaction data and historical user sentiment data in response to historical actions outputted to reduce dissatisfaction with the product to a plurality of users. This is illustrated at step 405. In embodiments, the system can utilize machine learning and/or deep learning, where algorithms or models can be generated by performing supervised, unsupervised, or semi-supervised training on historical user interaction data/user sentiment data for a plurality of users.”)
training the machine learning model based on the training subset by: inputting the training subset into the machine learning model (Roy Par. 19-“ n embodiments, the system utilizes analysis of user interaction data by the user with various IoT devices through the IoT network to reveal insights about the user's behavior toward the product (e.g., the IoT device itself, an application on the IoT device, a brand related to one or more IoT devices, etc.). The system uses various machine learning and/or artificial intelligence techniques to analyze the user interaction data to generate an initial usage pattern for the respective user with the given product.”);
receiving, from the machine learning model, predicted measures of future engagement of users with the application based on the inputting; comparing the predicted measures of future user engagement with actual engagement data in the validation subset;( Roy Par. 19; Par. 37; Par. 53- “The process 200 continues by comparing the second set of user interaction data and the second set of user sentiment data to the satisfaction threshold. This is illustrated at step 220. In embodiments, the system may use a scoring model when comparing the data. For example, the second set of user interaction data and the second set of user sentiment data may be provided a score and that score will be compared to the scores used to generate the satisfaction threshold (as described in FIG. 3). In some embodiments, the first set of user interaction data and the first set of user sentiment data may be compared directly to the second set of user interaction data and the second set of user sentiment data, respectively (e.g., without generating a score). The process 200 continues by determining if the satisfaction threshold has been exceeded.”);
Harris, Brown and Roy are directed to model analysis. It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have improve upon data analysis of Harris in view of Brown, as taught by Roy, by utilizing product interaction analysis with a reasonable expectation of success of arriving at the claimed invention. One of ordinary skill in the art would have been motivated to make the modification to the teachings of Harris in view of Brown with the motivation of improved customer relationship between the user and the product's maker, service provider, and/or brand ( Roy Par. 25).
Regarding Claim 10, Harris in view of Brown in further view of Roy teach The method of claim 1, further comprising: …
Harris teaches model analysis and the feature is expounded upon by Brown:
re-training the machine learning model based, at least in part, on the collected user feedback data, wherein retraining comprises at least: incorporating the collected user feedback data into the training dataset.(Brown Par. 65- “In some embodiments, and where such usage is permitted, the now classified data instances can be stored to the classified data repository, which can be used for further training of the trained model 908 by the training manager. In some embodiments the model will be continually trained as new data is available, but in other embodiments the models will be retrained periodically, such as once a day or week, depending upon factors such as the size of the data set or complexity of the model.”; Par. 70);
Harris and Brown are directed to model analysis. Brown improves upon the model analysis. It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have improve upon data analysis of Harris, as taught by Brown, by utilizing additional data analysis with a reasonable expectation of success of arriving at the claimed invention. One of ordinary skill in the art would have been motivated to make the modification to the teachings of Harris with the motivation of improving precision ( Brown Par. 29).
Harris in view of Brown fail to teach the following feature taught by Roy:
collecting user feedback data responsive to the application outputting the intervention, wherein the user feedback data comprises at least one of: user actions taken within the application following the intervention, and user engagement metrics with the application after the intervention; storing the collected user feedback data in a database; (Roy Par. 63-“The process 400 begins by analyzing, using machine learning, historical user interaction data and historical user sentiment data in response to historical actions outputted to reduce dissatisfaction with the product to a plurality of users. This is illustrated at step 405. In embodiments, the system can utilize machine learning and/or deep learning, where algorithms or models can be generated by performing supervised, unsupervised, or semi-supervised training on historical user interaction data/user sentiment data for a plurality of users.”; Roy Par. 17-“Embodiments of the present disclosure relate to a system for determining if a user (e.g., customer) is satisfied with a product and, if not, mitigating the reason(s) for dissatisfaction with the product before the customer relationship with the product's maker and/or service provider is strained.”; Par. 65-“The process 400 continues by implementing one of the set of historical actions when outputting the action (at step 235 of process 200) to reduce dissatisfaction of the user. This is illustrated at step 415. In this way, the system may continually predict the best strategies for obtaining a positive outcome on future occasions when outputting actions for reducing user dissatisfaction with the product.”; Par. 33-34)
Harris, Brown and Roy are directed to model analysis. It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have improve upon data analysis of Harris in view of Brown, as taught by Roy, by utilizing product interaction analysis with a reasonable expectation of success of arriving at the claimed invention. One of ordinary skill in the art would have been motivated to make the modification to the teachings of Harris in view of Brown with the motivation of improved customer relationship between the user and the product's maker, service provider, and/or brand ( Roy Par. 25).
Regarding Claim 11,
Harris teaches
A non-transitory computer-readable medium comprising memory with instructions encoded thereon, the instructions causing one or more processors to perform operations when executed, the instructions comprising instructions to: receive, based on user input into an application, a request for delivery of a first physical object (Harris Col 3 Ln 30-55 The access device 192 may present an interface including one or more control elements to receive delivery information status request inputs. The status request inputs may include one or more of item, delivery, or location information for which delivery status is requested.”; Col 10 Ln 54-64- FIG. 6 is a block diagram of an illustrative computing device that may implement one or more of the access control features described. The computing device 600 may implement the method or messaging shown in of FIG. 3 or 5 or generate the interface shown in FIG. 4. The computing device 600 can be a server or other computing device, and can comprise a processing unit 602, a delivery status processor 630, a network interface 604, a computer readable medium drive 606, an input/output device interface 608, and a memory 610. ) ;
detect a discrepancy wherein a second physical object is obtained instead of the first physical object (Harris col 7, In 40-62, The object detection may include selecting a computer vision model trained to detect delivery items from one or more images. The model may receive, as inputs, image data and, in some implementations, estimated dimensions of the delivered item. The model may provide as an output detection information indicating whether an object (e.g., a delivered item) is shown in the images provided to the model. The output information may include a detection result indicating a likelihood that an image shows an item associated with the delivery event, col 9-10- The analysis result may be used at block 530 to determine whether the item is represented in the monitoring data. The determination may compare an output value from the model to a detection threshold. );
input a first set of features of the first physical object and a second set of features of the second physical object into a machine learning model …(Harris col 7, In 40-62, The object detection may include selecting a computer vision model trained to detect delivery items from one or more images. The model may receive, as inputs, image data and, in some implementations, estimated dimensions of the delivered item. The model may provide as an output detection information indicating whether an object (e.g., a delivered item) is shown in the images provided to the model. The output information may include a detection result indicating a likelihood that an image shows an item associated with the delivery event - If the determination at block 530 is negative, the method 500 may proceed to block 535 to transmit the delivery event information for further review The review system may include sophisticated machine learning models (e.g., a vision model for image monitoring data- the failure may be due to an improper delivery (e.g., item left in an inappropriate location or unauthorized removal of the item from the physical location), col 9-10);
Harris fails to teach the following feature taught by Brown:
…wherein instructions to input further comprise instructions to: convert the features of the first physical object and the features of the second physical object into latent space embeddings; generate a similarity metric based on the latent space embeddings; and input the similarity metric into the machine learning model; (Brown Par. 23- “A computational analogue of these relative spaces involves vector space models. Vector models often find use in information retrieval and natural language processing as methods to encode information according to some basis decomposition that may be derived from keywords or phrases that capture semantic variance. Vector space methods provide an encoding scheme to convert information into points in a high-dimensional space where nearness in the space reflects factors such as relevance, relatedness, or semantic similarity. More generally, these vector spaces can form an embedding which is a mathematical structure that imposes a relationship among objects. When an embedding imposes a relationship between objects using distance it creates a metric space which induces a topology.; Par. 24-From a machine learning perspective, these methods can utilize supervised learning techniques in that “side-information” of similarity or neighbor labels are prescribed a priori. A learning metric process can be obtained that is unsupervised. To do this, the temporal correlations in visual sensory data can be leveraged by manipulating the data as a time-ordered sequence of images, where adjacent image pairs form positive training examples that exhibit both similarity of perceptual features and nearness of physical locality. Such a process can exploit temporal contiguity and constrain perceptual organization in both space and time. Such an organization might also facilitate an intuitive representation of temporal context. Par. 27 discloses “A goal of such a process can be to capture sensory relatedness such that similar features map to nearby points in a vector space embedding….The feature vectors, having a second size, can then be fed to an embedding network 406 as training data. Such a process of feature extraction followed by feature embedding produces a ‘relative space’ representation, or visualization 408, that is topologically consistent with the latent structure of the sensory information and draws a compelling analogy with the complementary nature of the visual cortex and the hippocampus.”; Par. 23-27”);
Harris and Brown are directed to model analysis. It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have improve upon data analysis of Harris, as taught by Brown, by utilizing additional data analysis with a reasonable expectation of success of arriving at the claimed invention. One of ordinary skill in the art would have been motivated to make the modification to the teachings of Harris with the motivation of improving precision ( Brown Par. 29).
Harris and Brown teach model analysis and the feature is expounded upon by Roy:
receive, as output from the machine learning model, a measure of predicted future engagement of the user with the application based on the discrepancy (Roy Par. 19; Par. 37; Par. 53- “The process 200 continues by comparing the second set of user interaction data and the second set of user sentiment data to the satisfaction threshold. This is illustrated at step 220. In embodiments, the system may use a scoring model when comparing the data. For example, the second set of user interaction data and the second set of user sentiment data may be provided a score and that score will be compared to the scores used to generate the satisfaction threshold (as described in FIG. 3). In some embodiments, the first set of user interaction data and the first set of user sentiment data may be compared directly to the second set of user interaction data and the second set of user sentiment data, respectively (e.g., without generating a score). The process 200 continues by determining if the satisfaction threshold has been exceeded.”);
and instruct the application to output an intervention based on the measure of predicted future engagement of the user. (Roy Par. 17-“Embodiments of the present disclosure relate to a system for determining if a user (e.g., customer) is satisfied with a product and, if not, mitigating the reason(s) for dissatisfaction with the product before the customer relationship with the product's maker and/or service provider is strained.”; Par. 65-“The process 400 continues by implementing one of the set of historical actions when outputting the action (at step 235 of process 200) to reduce dissatisfaction of the user. This is illustrated at step 415. In this way, the system may continually predict the best strategies for obtaining a positive outcome on future occasions when outputting actions for reducing user dissatisfaction with the product.”)
Harris, Brown and Roy are directed to model analysis. It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have improve upon data analysis of Harris in view of Brown, as taught by Roy, by utilizing product interaction analysis with a reasonable expectation of success of arriving at the claimed invention. One of ordinary skill in the art would have been motivated to make the modification to the teachings of Harris in view of Brown with the motivation of improved customer relationship between the user and the product's maker, service provider, and/or brand ( Roy Par. 25).
Regarding Claim 20,
Harris teaches
A computer system comprising: one or more processors; and a non-transitory computer readable storage medium storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising: receiving, based on user input into an application, a request for delivery of a first physical object (Harris Col 3 Ln 30-55 The access device 192 may present an interface including one or more control elements to receive delivery information status request inputs. The status request inputs may include one or more of item, delivery, or location information for which delivery status is requested.”; Col 10 Ln 54-64- FIG. 6 is a block diagram of an illustrative computing device that may implement one or more of the access control features described. The computing device 600 may implement the method or messaging shown in of FIG. 3 or 5 or generate the interface shown in FIG. 4. The computing device 600 can be a server or other computing device, and can comprise a processing unit 602, a delivery status processor 630, a network interface 604, a computer readable medium drive 606, an input/output device interface 608, and a memory 610. ) ;
receiving information from a scanning system identifying a second physical object as satisfying the request for the first physical object; (Harris Col 6Ln46-66-The delivery agent 120 may interact with one or more interfaces presented via the delivery device 122. The interfaces may include control elements to receive input or adjust a function of the delivery device 122. For example, the delivery device 122 may include a camera or other optical scanning element. The delivery device 122 may activate the camera to scan an item 124 to be delivered. For example, the item 124 may include a scannable code or other detectable indicator. The scannable code may encode or indicate an identifier for the item or delivery information related to the item such as address, unique identifier of a monitoring device 150, or other information used in identifying or delivering the item 124.; Col 8 Ln 5-25- For example, some physical locations may have ornamental features that are mistaken for boxes. In such instances, the delivery information system 200 may incorrectly associate the delivery event with monitoring data associated with a non-delivery monitoring event. The feedback can be included during automatic identification to select a vision model that is adapted to more accurately identify the portions of the monitoring data associated with delivery events. For example, a general vision model may be used which can quickly identify portions for a majority of the instances of monitoring data. However, for physical locations where negative feedback is received, a slower but more accurate model may be used to process the monitoring data generated by a monitoring device associated with the location.; Col 9-10- At block 510, the delivery information system 200 may determine whether the delivery event is associated with a monitoring threshold. The monitoring threshold may identify a size of the portion of monitoring data for presenting the delivery event. For example, for a delivery event it may be desirable to include two minutes of monitoring data whereas for a security event it may be desirable to show fifteen minutes or more of the monitoring data. The threshold may be defined using a delivery information configuration stored in memory. In some implementations, the threshold may be specific to one or more of a user, a physical location, or a monitoring device.”)
detecting a discrepancy based on detecting that the second physical object is not a same physical object as the first physical object (Harris col 7, In 40-62, The object detection may include selecting a computer vision model trained to detect delivery items from one or more images. The model may receive, as inputs, image data and, in some implementations, estimated dimensions of the delivered item. The model may provide as an output detection information indicating whether an object (e.g., a delivered item) is shown in the images provided to the model. The output information may include a detection result indicating a likelihood that an image shows an item associated with the delivery event, col 9-10- The analysis result may be used at block 530 to determine whether the item is represented in the monitoring data. The determination may compare an output value from the model to a detection threshold.; Col 8 Ln 5-2; col 9-10- The analysis result may be used at block 530 to determine whether the item is represented in the monitoring data. The determination may compare an output value from the model to a detection threshold. );
inputting a first set of features of the first physical object and a second set of features of the second physical object into a machine learning model… (Harris col 7, In 40-62, The object detection may include selecting a computer vision model trained to detect delivery items from one or more images. The model may receive, as inputs, image data and, in some implementations, estimated dimensions of the delivered item. The model may provide as an output detection information indicating whether an object (e.g., a delivered item) is shown in the images provided to the model. The output information may include a detection result indicating a likelihood that an image shows an item associated with the delivery event - If the determination at block 530 is negative, the method 500 may proceed to block 535 to transmit the delivery event information for further review The review system may include sophisticated machine learning models (e.g., a vision model for image monitoring data- the failure may be due to an improper delivery (e.g., item left in an inappropriate location or unauthorized removal of the item from the physical location), col 9-10);
Harris in view of Roy fail to teach the following feature taught by Brown:
…wherein instructions to input further comprise instructions to: convert the features of the first physical object and the features of the second physical object into latent space embeddings; generate a similarity metric based on the latent space embeddings; and input the similarity metric into the machine learning model; (Brown Par. 23- “A computational analogue of these relative spaces involves vector space models. Vector models often find use in information retrieval and natural language processing as methods to encode information according to some basis decomposition that may be derived from keywords or phrases that capture semantic variance. Vector space methods provide an encoding scheme to convert information into points in a high-dimensional space where nearness in the space reflects factors such as relevance, relatedness, or semantic similarity. More generally, these vector spaces can form an embedding which is a mathematical structure that imposes a relationship among objects. When an embedding imposes a relationship between objects using distance it creates a metric space which induces a topology.; Par. 23-27”);
Harris and Roy are directed to model analysis. Brown improves upon the data analysis. It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have improve upon data analysis of Harris in view of Roy, as taught by Brown, by utilizing additional data analysis with a reasonable expectation of success of arriving at the claimed invention. One of ordinary skill in the art would have been motivated to make the modification to the teachings of Harris in view of Roy with the motivation of improving precision ( Brown Par. 29).
Harris in view of Brown teach product analysis and the feature is expounded upon by Roy:
receiving, as output from the machine learning model, a measure of predicted future engagement of the user with the application based on the discrepancy (Roy Par. 19; Par. 37; Par. 53- “The process 200 continues by comparing the second set of user interaction data and the second set of user sentiment data to the satisfaction threshold. This is illustrated at step 220. In embodiments, the system may use a scoring model when comparing the data. For example, the second set of user interaction data and the second set of user sentiment data may be provided a score and that score will be compared to the scores used to generate the satisfaction threshold (as described in FIG. 3). In some embodiments, the first set of user interaction data and the first set of user sentiment data may be compared directly to the second set of user interaction data and the second set of user sentiment data, respectively (e.g., without generating a score). The process 200 continues by determining if the satisfaction threshold has been exceeded.”);
and instructing the application to output an intervention based on the measure of predicted future engagement of the user. (Roy Par. 17-“Embodiments of the present disclosure relate to a system for determining if a user (e.g., customer) is satisfied with a product and, if not, mitigating the reason(s) for dissatisfaction with the product before the customer relationship with the product's maker and/or service provider is strained.”; Par. 65-“The process 400 continues by implementing one of the set of historical actions when outputting the action (at step 235 of process 200) to reduce dissatisfaction of the user. This is illustrated at step 415. In this way, the system may continually predict the best strategies for obtaining a positive outcome on future occasions when outputting actions for reducing user dissatisfaction with the product.”)
Harris, Brown and Roy are directed to model analysis. It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have improve upon data analysis of Harris in view of Brown, as taught by Roy, by utilizing product interaction analysis with a reasonable expectation of success of arriving at the claimed invention. One of ordinary skill in the art would have been motivated to make the modification to the teachings of Harris in view of Brown with the motivation of improved customer relationship between the user and the product's maker, service provider, and/or brand ( Roy Par. 25).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure: US Publication No. 20240152955A1 to Mao et al.- Abstract-“ Methods, systems, and computer programs are presented for eliminating bias while training an ML model using training data that includes past experimental data. One method includes accessing experiment results, for A/B testing of a first model, that comprise information regarding engagement with a first set of items presented to users, each item being presented within an ordered list of results. A position bias is calculated for positions within the ordered list of results where the items were presented. A machine-learning program is trained to obtain a second model using a training set comprising values for features that include the calculated position bias. The method includes detecting a second set of items to be ranked for presentation to a first user, and calculates, using the second model, a relevance score for the second set of items, which are ranked based on the respective relevance score and presented on a display.”
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 Chesiree Walton, whose telephone number is (571) 272-5219. The examiner can normally be reached from Monday to Friday between 8 AM and 5 PM. If any attempt to reach the examiner by telephone is unsuccessful, the examiner’s supervisor, Patricia Munson, can be reached at (571) 270-5396. The fax telephone numbers for this group are either (571) 273-8300 or (703) 872-9326 (for official communications including After Final communications labeled “Box AF”).
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://pair-direct.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 an in-person 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.
Sincerely,
/CHESIREE A WALTON/Examiner, Art Unit 3624
Prosecution Insights