Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 02/26/2026 has been entered.
Response to Arguments
Regarding objections to claims 1 and 13 because of informalities, amendments to the claims have overcome the objections, which are withdrawn.
Regarding the rejection of claims under 35 U.S.C. 103, Applicant’s arguments are directed towards amendments to claims that have not been previously examined (“Yoshida does not teach or suggest ‘determining a direction of a change in the target variable data, wherein the direction of the change in the target variable data is determined to indicate an increase of the process yield or the return on investment’ of Claim 1”), or are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument (“Yoshida does not teach the operation of ‘learning control variable data for
optimizing the target variable data according to the determined direction of change in the target variable data’ of Claim 1”).
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 9-13, and 16-17 rejected under 35 U.S.C. 103 over Achin et al., US Pre-Grant Publication No. 2018/0060738 (hereafter Achin) in view of Bromfield et al., US Pre-Grant Publication No. 2021/0048016 (hereafter Bromfield) and Yoshida et al., US Pre-Grant Publication No. 2017/0003676 (hereafter Yoshida).
Regarding claim 1 and analogous claims 12-13:
Achin teaches:
“A method of predicting, controlling, and describing time series data based on automatic learning, the method performed by an electronic device comprising a memory storing computer-readable instructions and a hardware processor configured to execute the instructions to perform the method comprising”: Achin, paragraph 0060, “In some embodiments, predictive modeling system 100 may support time-series prediction problems [predicting, controlling, and describing time series data based on automatic learning] (e.g., uni-dimensional or multi-dimensional time-series prediction problems)”; Achin, paragraph 0026, “According to another aspect of the present disclosure, a predictive modeling apparatus is provided, including a memory configured to store processor-executable instructions; and a processor configured to execute the processor executable instructions, wherein executing the processor executable instructions causes the apparatus to perform steps [performed by an electronic device comprising a memory storing computer-readable instructions and a hardware processor configured to execute the instructions to perform the method] including […]”; Achin, paragraph 0342, “In this respect, some embodiments may be embodied as a computer readable medium ( or multiple computer readable media) (e.g., a computer memory, one or more floppy discs, compact discs, optical discs, magnetic tapes, flash memories, circuit configurations in Field Programmable Gate Arrays or other semiconductor devices, or other tangible computer storage medium) encoded with one or more programs that, when executed on one or more computers or other processors, perform methods that implement the various embodiments discussed above. The computer readable medium or media may be non-transitory.”
“training a plurality of time series data prediction models according to conditions for the respective models”: Achin, paragraph 0060, “In some embodiments, predictive modeling system 100 may support time-series prediction problems [time series data prediction models] (e.g., uni-dimensional or multi-dimensional time-series prediction problems).”; Achin, paragraph 0102, “In some embodiments, testing a predictive model includes cross-validating the model using different folds of training datasets associated with the prediction problem [training]. In some embodiments, the execution of the modeling procedures includes the testing of the generated models. In some embodiments, the testing of the generated models is performed separately from the execution of the modeling procedures”; Achin, paragraph 0023, “In some embodiments, the method further includes, prior to performing the plurality of second modeling procedures: selecting the second modeling procedures based on suitabilities of the selected modeling procedures for the initial prediction problem, wherein a suitability of the particular predictive modeling procedure for the initial prediction problem is determined based, at least in part, on characteristics of one or more particular features of the initial dataset having high model-specific predictive values for the particular predictive modeling procedure [according to conditions for the respective models].”
“determining optimal models among the trained time series data prediction models, wherein each of the optimal models meets a predetermined condition”: Achin, paragraph 0023, “In some embodiments, the method further includes, prior to performing the plurality of second modeling procedures: selecting the second modeling procedures based on suitabilities of the selected modeling procedures for the initial prediction problem, wherein a suitability of the particular predictive modeling procedure for the initial prediction problem is determined based, at least in part, on characteristics of one or more particular features of the initial dataset having high model-specific predictive values for the particular predictive modeling procedure [determining optimal models among the trained time series data prediction models, wherein each of the optimal models meets a predetermined condition].”
“generating a final model by combining the optimal models”: Achin, paragraph 0024, “In some embodiments, the method further includes: generating a blended predictive model by combining two or more of the generated predictive models [generating a final model by combining the optimal models]; and evaluating the blended predictive model.”
“receiving target variable data for predicting time series data, inputting the target variable data to the final model, and generating target variable prediction data corresponding to the target variable data by using the final model”: Achin, paragraph 0004, “Based on the outcomes predicted by the predictive models, organizations can make decisions, adjust processes, or take other actions. For example, an insurance company might seek to build a predictive model that more accurately forecasts future claims, or a predictive model that predicts when policyholders are considering switching to competing insurers. An automobile manufacturer might seek to build a predictive model that more accurately forecasts demand for new car models. A fire department might seek to build a predictive model that forecasts days with high fire danger, or predicts which structures are endangered by a fire [receiving target variable data for predicting time series data, inputting the target variable data to the final model, and generating target variable prediction data corresponding to the target variable data by using the final model].”
“wherein the plurality of time series data prediction models comprise at least one of statistical-based prediction models and deep learning-based prediction models”: Achin, paragraph 0005, “Machine-learning techniques (e.g., supervised statistical-learning techniques) [statistical-based prediction models] may be used to generate a predictive model from a dataset that includes previously recorded observations of at least two variables.”
(bold only) “wherein the adjustment direction of the optimized control variable data includes an optimal value search direction and an optimal value search time of control variable data”: Achin, paragraph 0237, “However, predictive modeling system 100 doesn't force the user to ‘fire-and-forget’, i.e., stop his own engagement with the problem until receiving a notification. In fact, it may encourage him to continue exploring the dataset and review preliminary results as soon as they arrive. Such additional exploration or initial insight might inspire him to change the model-building parameters "in-flight". The system may then process the requested changes and reallocate processing tasks. The predictive modeling system 100 may allow this request-and-revise dynamic continuously throughout the user's session [an optimal value search time of control variable data, interpreted as including an interaction with a user regarding the period for consideration of a control optimization process].”
Achin does not explicitly teach:
“receiving control variable data, inputting the control variable data to the final model, and generating control variable prediction data corresponding to the control variable data by using the final model”
wherein the target variable data includes at least one of data of process yield and data of return on investment over time”
and the control variable data is data that determines a direction of a change in the target variable prediction data”
“providing a prediction result and a control method for the times-series data based on the target variable prediction data and the control variable prediction data”
“wherein the providing of the prediction result and the control method for the times-series data comprises: determining a direction of a change in the target variable data”
“wherein the direction of the change in the target variable data is determined to indicate an increase of the process yield or the return on investment”
“learning control variable data for optimizing the target variable data according to the determined direction of change in the target variable data”
“providing a user with an adjustment direction of the optimized control variable data”
(bold only) “wherein the adjustment direction of the optimized control variable data includes an optimal value search direction and an optimal value search time of control variable data”
Bromfield teaches:
“receiving control variable data, inputting the control variable data to the final model, and generating control variable prediction data corresponding to the control variable data by using the final model”: Bromfield, paragraph 0024, “Industrial automation environment 110 also includes a plurality of sensors 132 and 134 which are configured to transmit current environmental data to universal compressor control 120 over link 115 or other links (not shown). This environmental data may include such data as: temperatures, air pressures, air flows, motion, vibration, and the like [receiving control variable data]”; Bromfield, paragraph 0028, “Optimization module 126 is coupled with control module 122 and analysis module 124 and configured to process the performance data, current environment data, guide vane weights, and target system air pressure to determine control settings for each of the plurality of compressors [inputting the control variable data to the final model, and generating control variable prediction data corresponding to the control variable data by using the final model].”
“wherein the target variable data includes at least one of data of process yield and data of return on investment over time”: Bromfield, paragraph 0029, “In some embodiments, optimization module 126 Is also configured to calculate an efficiency for each of compressors 140, 142, 144, and 146 based on the performance data and guide vane weight, and prioritize more efficient compressors over less efficient compressors [wherein the target variable data includes at least one of data of process yield and data of return on investment over time] while processing the performance data, current environment data, guide vane weights, and target system air pressure to determine control settings for each of compressors 140, 142, 144 and 146. In some embodiments, optimization module also processes a model of compressed air distribution system in determining control settings for the compressors. This model may be very complex and includes data such as the physical structure of compressed air distribution system 130 which may range over several miles, and have different physical and performance characteristics across its length.”
“and the control variable data is data that determines a direction of a change in the target variable prediction data”: Bromfield, paragraph 0024, “Industrial automation environment 110 also includes a plurality of sensors 132 and 134 which are configured to transmit current environmental data to universal compressor control 120 over link 115 or other links (not shown). This environmental data may include such data as: temperatures, air pressures [target variable prediction data], air flows, motion, vibration, and the like”; Bromfield, paragraph 0028, “Optimization module 126 is coupled with control module 122 and analysis module 124 and configured to process the performance data, current environment data, guide vane weights, and target system air pressure to determine control settings for each of the plurality of compressors”; Bromfield, paragraph 0028, “Optimization module 126 is coupled with control module 122 and analysis module 124 and configured to process the performance data, current environment data, guide vane weights, and target system air pressure to determine control settings for each of the plurality of compressors [control data is used to adjust a measured value towards a target, hence, the control variable data is data that determines a direction of a change in the target variable prediction data].”
“providing a prediction result and a control method for the times-series data based on the target variable prediction data and the control variable prediction data, wherein the providing of the prediction result and the control method for the times-series data comprises: determining a direction of a change in the target variable data”: Bromfield, paragraph 0028, “Optimization module 126 is coupled with control module 122 and analysis module 124 and configured to process the performance data, current environment data, guide vane weights, and target system air pressure [target variable prediction data] to determine control settings for each of the plurality of compressors. Control module 122 then transmits the control settings [control variable prediction data] to compressor controllers 141, 143, 145, and 147 over link 115 [providing a prediction result and a control method for the times-series data based on the target variable prediction data and the control variable prediction data]”; Bromfield, paragraph 0028, “Optimization module 126 is coupled with control module 122 and analysis module 124 and configured to process the performance data, current environment data, guide vane weights, and target system air pressure to determine control settings [providing of the prediction result and the control method for the times-series data] for each of the plurality of compressors [control data is used to adjust a measured value towards a target, hence, determining a direction of a change in the target variable data].”
“wherein the direction of the change in the target variable data is determined to indicate an increase of the process yield or the return on investment”: Bromfield, paragraph 0028, “Optimization module 126 is coupled with control module 122 and analysis module 124 and configured to process the performance data, current environment data, guide vane weights, and target system air pressure to determine control settings for each of the plurality of compressors”; Bromfield, paragraph 0029, “In some embodiments, optimization module 126 Is also configured to calculate an efficiency for each of compressors 140, 142, 144, and 146 based on the performance data and guide vane weight, and prioritize more efficient compressors over less efficient compressors while processing the performance data, current environment data, guide vane weights, and target system air pressure to determine control settings for each of compressors 140, 142, 144 and 146 [adjusting control data such that a monitored value is increased or lowered to meet a target, where such adjustments result in higher efficiency, hence wherein the direction of the change in the target variable data is determined to indicate an increase of the process yield or the return on investment]. In some embodiments, optimization module also processes a model of compressed air distribution system in determining control settings for the compressors. This model may be very complex and includes data such as the physical structure of compressed air distribution system 130 which may range over several miles, and have different physical and performance characteristics across its length.”
“learning control variable data for optimizing the target variable data according to the determined direction of change in the target variable data”: Bromfield, paragraph 0028, “Optimization module 126 is coupled with control module 122 and analysis module 124 and configured to process the performance data, current environment data, guide vane weights, and target system air pressure to determine control settings for each of the plurality of compressors”; Bromfield, paragraph 0029, “In some embodiments, optimization module 126 Is also configured to calculate an efficiency for each of compressors 140, 142, 144, and 146 based on the performance data and guide vane weight, and prioritize more efficient compressors over less efficient compressors while processing the performance data, current environment data, guide vane weights, and target system air pressure [adjusting control data such that a monitored value is increased or lowered to meet a target, hence according to the determined direction of change in the target variable data] to determine control settings for each of compressors 140, 142, 144 and 146. In some embodiments, optimization module 126 also processes a model of compressed air distribution system in determining control settings for the compressors. This model may be very complex and includes data such as the physical structure of compressed air distribution system 130 which may range over several miles, and have different physical and performance characteristics across its length”; Bromfield, paragraph 0031, “Predictive/machine learning module 128 is coupled with control module 122, analysis module 124, and optimization module 126, and is configured to monitor the performance data from compressor controllers 141, 143, 145, and 147 and the current environment data from sensors 132 and 134 over a period of time, such as a month, and to process the monitored performance data and current environment data to predict future control settings for the plurality of compressors [learning control variable data for optimizing the target variable data].”
Bromfield and Achin are analogous arts as they are both related to data modelling. It would have been obvious to a person having ordinary skill in the art prior to the effective filing date of the claimed invention to have combined the control system of Bromfield with the teachings of Achin to arrive at the present invention, in order to use the combined model to optimize a resulting process, as stated in Bromfield, paragraph 0032, “This predictive function allows universal compressor controller 120 to anticipate recurring needs for extra pressure, potential failures, cyclic changes in efficiency, and the like.”
Yoshida teaches:
“providing a user with an adjustment direction of the optimized control variable data”: Yoshida, paragraph 0053, “The external output interface 14 outputs to an external output device 12 the optimized result of the objective function outputted from the control parameter optimizing section 2 of the control parameter optimizing system 100. The external output device 12 is configured by a device that has a screen display capability, such as a personal computer (PC) monitor. In this manner, the operator can verify the optimized result of the objective function via the external output device 12 [providing a user with an adjustment direction of the optimized control variable data].”
(bold only) “wherein the adjustment direction of the optimized control variable data includes an optimal value search direction and an optimal value search time of control variable data”: Yoshida, paragraph 0023, “First, the optimization control parameter selecting section 7 detects from the control logic modules 21 to 24 a signal corresponding to the objective function (assumed to be the objective function Al in this example) set by the objective function setting section 1. One way of detecting such a control signal is by searching the control logic modules for the signal having the name of a character string identical or similar to the character string constituting the name of the objective function. In this example, a signal Al with a character string name that matches the character string ‘Al’ of the objective function Al is detected from the control logic module 21. Preferably, the detected signal corresponding to the objective function may be displayed on an external monitor, for example, so that the operator can verify the detected signal. If a plurality of signals have been detected, the operator may be prompted to select the appropriate signal on the monitor [the adjustment direction of the optimized control variable data includes an optimal value search direction, interpreted as including an interaction with a user regarding data related to control optimization]. In any case, arrangements can be made to let the signal corresponding to the objective function be detected in an interactive manner.”
Yoshida and Achin are analogous arts as they are both related to predictive machine learning modelling. It would have been obvious to a person having ordinary skill in the art prior to the effective filing date of the claimed invention to have combined the control optimization of Yoshida with the teachings of Achin to arrive at the present invention, in order to provide for the optimization of the operation of a system, as stated in Yoshida, paragraph 0009, “According to the present invention, it is possible to optimize the operation control of an existing plant regardless of the plant type or control panel specifications and without recourse to modifying the control panel or equipment of the plant.”
Regarding claim 9:
Achin as modified by Bromfield and Yoshida teaches “the method of claim 1.”
Achin further teaches “wherein the training comprises training the plurality of time series data prediction models a predetermined number of times according to the conditions for the respective models”: Achin, paragraph 0105, “In some embodiments, selecting the predictive model for the prediction problem may comprise iteratively selecting a subset of the predictive models and training the
selected predictive models on larger or different portions of the dataset. This iterative process may continue until a predictive model is selected for the prediction problem or until the processing resources budgeted for generating the predictive model are exhausted [a fixed budget setting a limit for training iterations, hence, training the plurality of time series data prediction models a predetermined number of times according to the conditions for the respective models].”
Regarding claim 10:
Achin as modified by Bromfield and Yoshida teaches “the method of claim 1.”
Achin further teaches “evaluating prediction performance of the final model; and updating the final model, when the prediction performance of the final model decreases below a predetermined threshold”: Achin, paragraphs 0105-1016, “In some embodiments, selecting the predictive model for the prediction problem may comprise iteratively selecting a subset of the predictive models and training the selected predictive models on larger or different portions of the dataset. This iterative process may continue until a predictive model is selected [updating the final model] for the prediction problem or until the processing resources budgeted for generating the predictive model are exhausted. Selecting a subset of predictive models may comprise selecting a fraction of the predictive models with the highest scores, selecting all models having scores that exceed a threshold score [when the prediction performance of the final model decreases below a predetermined threshold], selecting all models having scores within a specified range of the score of the highest-scoring model, or selecting any other suitable group of models.”
Regarding claim 11:
Achin as modified by Bromfield and Yoshida teaches “the method of claim 1.”
Achin further teaches “updating the final model according to a predetermined interval”: Achin, paragraph 0126, “Creating derived features by interpreting and transforming the original features can increase the dimensionality of the original dataset. The predictive modeling system 100 may apply dimension reduction techniques, which may counter the increase in the dataset's dimensionality. However, some modeling techniques are more sensitive to dimensionality than others. Also, different dimension reduction techniques tend to work better with some modeling techniques than others. In some embodiments, predictive modeling system 100 maintains metadata describing these interactions. The system 100 may systematically evaluate various combinations of dimension reduction techniques and modeling techniques, prioritizing the combinations that the metadata indicate are most likely to succeed. The system may further update this metadata based on the empirical performance of the combinations over time and incorporate new dimension reduction techniques as they are discovered [updating the final model according to a predetermined interval, interpreted as including a systematic regular update of model parameters].”
Regarding claim 16:
Achin as modified by Bromfield and Yoshida teaches “the method of claim 1.”
Yoshida further teaches “wherein the electronic device further comprises a black box that is a model configured to output a result value corresponding to independent variables that are the control variable data”: Yoshida, Fig. 3,
PNG
media_image1.png
715
1095
media_image1.png
Greyscale
[showing plant model 3, which output control parameter data to the control parameter optimization section 2, hence, wherein the electronic device further comprises a black box that is a model configured to output a result value corresponding to independent variables that are the control variable data].
Yoshida and Achin are combinable for the rationale given under claim 1.
Regarding claim 17:
Achin as modified by Bromfield and Yoshida teaches “the method of claim 16.”
Yoshida further teaches:
“wherein the providing of the user with the adjustment direction of the optimized control variable data comprises: selecting a moving direction from an origin of the independent variables according to a correlation between the independent variables and a dependent variable that the final model learned”: Yoshida, paragraph 0029, “From the related control parameters extracted in the above-described extraction steps, the optimization control parameter selecting section 7 selects as the optimization control parameter one or a plurality of related control parameters having high sensitivity to the objective function. The sensitivity of a related control parameter to an objective function is defined, for example, by the ratio of the amount of change in the objective function to the amount of change in the related control parameter [the target variable and the control variable are related by ratio, therefore they move in the same direction, hence, selecting a moving direction from an origin of the independent variables according to a correlation between the independent variables and a dependent variable that the final model learned].”
“inputting a stored optimal value search result to the black box model when a preset search end time is reached”: Yoshida, paragraph 0031-0033, “The optimization control parameter adjusting section 8 first sets a predetermined value to the optimization control parameter selected by the optimization control parameter selecting section 7. The optimization control parameter adjusting section 8 then inputs the optimization control parameter to the plant model 3. The plant model 3 calculates the objective function based on the value of the optimization control parameter inputted from the optimization control parameter adjusting section 8 [inputting a stored optimal value search result to the black box model, interpreted as including using an optimized parameter value] using a control model 9 and a physical model 10 (both to be discussed later). The optimization control parameter adjusting section 8 adjusts the value of the optimization control parameter in such a manner as to minimize the difference between the calculated value of the objective function outputted from the plant model 3 and a predetermined target value. The optimization control parameter adjusting section 8 adjusts the optimization control parameter value by performing the above-described adjustment steps once or a number of times [when a preset search end time is reached, interpreted as including processes that process for a specified number of iterations]. An existing optimization algorithm such as the multi-objective evolutionary algorithm or the successive quadratic programming method may be used in adjusting the value of the optimization control parameter.”
“determining a search time at which a response of the black box model is optimal as the optimal search time”: Yoshida, paragraph 0031-0033, “The optimization control parameter adjusting section 8 first sets a predetermined value to the optimization control parameter selected by the optimization control parameter selecting section 7. The optimization control parameter adjusting section 8 then inputs the optimization control parameter to the plant model 3. The plant model 3 calculates the objective function based on the value of the optimization control parameter inputted from the optimization control parameter adjusting section 8 using a control model 9 and a physical model 10 (both to be discussed later). The optimization control parameter adjusting section 8 adjusts the value of the optimization control parameter in such a manner as to minimize the difference between the calculated value of the objective function outputted from the plant model 3 and a predetermined target value. The optimization control parameter adjusting section 8 adjusts the optimization control parameter value by performing the above-described adjustment steps once or a number of times [determining a search time at which a response of the black box model is optimal as the optimal search time, interpreted as including determination of the number of iterations to continue an optimization process]. An existing optimization algorithm such as the multi-objective evolutionary algorithm or the successive quadratic programming method may be used in adjusting the value of the optimization control parameter.”
“determining the optimal value search result at the optimal search time as a guide”: Yoshida, paragraph 0032, “The optimization control parameter adjusting section 8 adjusts the value of the optimization control parameter in such a manner as to minimize the difference between the calculated value of the objective function outputted from the plant model 3 and a predetermined target value [determining the optimal value search result at the optimal search time as a guide].”
Yoshida and Achin are combinable for the rationale given under claim 1.
Claim 18 rejected under 35 U.S.C. 103 over Achin as modified by Bromfield and Yoshida in view of Miah, “Algorithms – Min and Max,” 2017, https://medium.com/killingmeswiftly/algorithms-min-and-max-31917295f6b0 (hereafter Miah).
Achin as modified by Bromfield and Yoshida teaches “the method of claim 17.”
Achin further teaches (bold only) “wherein the selecting of the moving direction from the origin of the independent variables according to the correlation between the independent variables and the dependent variable that the final model learned comprises: when a response of the final model is improved in a search value obtained by searching the independent variables one time in an arbitrary direction from the origin, proceeding with conducting a search for a next optimal value, and storing the search value, and when the response of the final model is not improved in the search value, dismissing the search value, and returning the independent variables to a previous position”: Achin, paragraph 0024, “In some embodiments, the method further includes: generating a blended predictive model by combining two or more of the generated predictive models [a final model]; and evaluating the blended predictive model.”
Yoshida further teaches “wherein the selecting of the moving direction from the origin of the independent variables according to the correlation between the independent variables and the dependent variable that the final model learned comprises: when a response of the final model is improved in a search value obtained by searching the independent variables one time in an arbitrary direction from the origin, proceeding with conducting a search for a next optimal value, and storing the search value, and when the response of the final model is not improved in the search value, dismissing the search value, and returning the independent variables to a previous position”: Yoshida, paragraph 0029, “From the related control parameters extracted in the above-described extraction steps, the optimization control parameter selecting section 7 selects as the optimization control parameter one or a plurality of related control parameters having high sensitivity to the objective function. The sensitivity of a related control parameter to an objective function is defined, for example, by the ratio of the amount of change in the objective function to the amount of change in the related control parameter [the target variable and the control variable are related by ratio, therefore they move in the same direction, hence, selecting a moving direction from an origin of the independent variables according to a correlation between the independent variables and a dependent variable that the final model learned].”
Yoshida and Achin are combinable for the rationale given under claim 1.
Achin as modified by Bromfield and Yoshida does not explicitly teach (bold only) “wherein the selecting of the moving direction from the origin of the independent variables according to the correlation between the independent variables and the dependent variable that the final model learned comprises: when a response of the final model is improved in a search value obtained by searching the independent variables one time in an arbitrary direction from the origin, proceeding with conducting a search for a next optimal value, and storing the search value, and when the response of the final model is not improved in the search value, dismissing the search value, and returning the independent variables to a previous position.”
Miah teaches (bold only) “wherein the selecting of the moving direction from the origin of the independent variables according to the correlation between the independent variables and the dependent variable that the final model learned comprises: when a response of the final model is improved in a search value obtained by searching the independent variables one time in an arbitrary direction from the origin, proceeding with conducting a search for a next optimal value, and storing the search value, and when the response of the final model is not improved in the search value, dismissing the search value, and returning the independent variables to a previous position”: Miah, page 2, “Then we need to create placeholder variables that will contain the minimum and maximum values for us. Currently, set both of these to the first element in the array. We will be using the first element in our array as the reference for both the maximum and minimum values [a search value obtained by searching the independent variables one time in an arbitrary direction from the origin, arbitrary direction interpreted as any ordered processing of the search data], don’t worry these values are subject to change.
PNG
media_image2.png
62
356
media_image2.png
Greyscale
Now we will delve into our for-loop [proceeding with conducting a search for a next optimal value], but before we do let’s really think about what we are trying to accomplish with these variables we created. We are going to compare all the elements in our for loop with our variables min and max. Whichever element is smaller should become the new min value and whichever element is larger should become the larger max value.
PNG
media_image3.png
157
283
media_image3.png
Greyscale
[showing that the value of max is set to number when number is larger (improved), but max is unchanged when number is not an improvement, hence, storing the search value, and when the response … is not improved in the search value, dismissing the search value, and returning the independent variables to a previous position].
Miah and Achin as modified by Yoshida are analogous arts as the technique of Miah can be directly applied to optimization as performed in Achin and Yoshida. It would have been obvious to a person having ordinary skill in the art prior to the effective filing date of the claimed invention to have combined the maximization algorithm of Miah with the teachings of Achin as modified by Yoshida to arrive at the present invention, in order to identify the optimal value in a set of values, as stated in Miah, paragraph 1, “In today’s blog post, let’s go over a recent algorithm problem I solved without any helper/high order functions. I was told to find the minimum and maximum value in an array without the functions min and max.”
Claim 19 rejected under 35 U.S.C. 103 over Achin as modified by Bromfield and Yoshida in view of Wood et al., US Pre-Grant Publication No. 2015/0212974 (hereafter Wood).
Achin as modified by Bromfield and Yoshida teaches “the method of claim 1.”
Achin as modified by Bromfield and Yoshida does not explicitly teach “wherein the statistical-based prediction models comprise Least Absolute Shrinkage and Selection Operator (LASSO), Autoregressive Integrated Moving Average (ARIMA), and Extreme Gradient Boosting (XGBoost), and the deep learning-based prediction models comprise Fully Convolutional Network/Convolutional Neural Network (FCN/CNN), Long Short-Term Memory (LSTM), Long Short-Term Memory-Convolutional Neural Network (LSTM-CNN), Spatial Temporal aware Graph Convolutional Neural Network (STGCN), Dual Stage Attention-based Recurrent Neural Network (DA-RNN), and Deep Supervision-based Simple Attention Network (DSANet).”
Wood teaches “wherein the statistical-based prediction models comprise Least Absolute Shrinkage and Selection Operator (LASSO), Autoregressive Integrated Moving Average (ARIMA), and Extreme Gradient Boosting (XGBoost), and the deep learning-based prediction models comprise Fully Convolutional Network/Convolutional Neural Network (FCN/CNN), Long Short-Term Memory (LSTM), Long Short-Term Memory-Convolutional Neural Network (LSTM-CNN), Spatial Temporal aware Graph Convolutional Neural Network (STGCN), Dual Stage Attention-based Recurrent Neural Network (DA-RNN), and Deep Supervision-based Simple Attention Network (DSANet)”: Wood, paragraph 0003, “Auto-Regressive Integrated Moving-Average (‘ARIMA’) [Autoregressive Integrated Moving Average (ARIMA)] is a class of statistical models [statistical-based prediction models] used for modeling time-series data and forecasting future values of the time-series. Such modeling and forecasting can then be used for predicting events in the future and taking proactive actions and/or for detecting abnormal trend.”
Wood and Achin are analogous arts as they are both related to modelling of time-series data. It would have been obvious to a person having ordinary skill in the art prior to the effective filing date of the claimed invention to have combined the using of ARIMA from Wood with the teachings of Achin to arrive at the present invention, in order to make predictions from the time-series data and act upon then, as stated in Wood, paragraph 0003, “Auto-Regressive Integrated Moving-Average (‘ARIMA’) is a class of statistical models used for modeling time-series data and forecasting future values of the time-series. Such modeling and forecasting can then be used for predicting events in the future and taking proactive actions and/or for detecting abnormal trend.”
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Ineke et al., US Pre-Grant Publication No. 2007/0143197, discloses methods of predicting the direction of a market value for a financial asset.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to VINCENT SPRAUL whose telephone number is (703)756-1511. The examiner can normally be reached M-F 9:00 am - 5:00 pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, MICHAEL HUNTLEY can be reached at (303) 297-4307. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/VAS/Examiner, Art Unit 2129
/MICHAEL J HUNTLEY/Supervisory Patent Examiner, Art Unit 2129