Prosecution Insights
Last updated: November 04, 2026
Application No. 17/919,386

METHOD, DEVICE AND APPARATUS FOR EXECUTION OF AUTOMATED MACHINE LEARNING PROCESS

Non-Final OA §102§103
Filed
Oct 17, 2022
Priority
Apr 17, 2020 — CN 202010307807.5 +1 more
Examiner
MEYER, JACQUELINE CHRISTINE
Art Unit
2144
Tech Center
2100 — Computer Architecture & Software
Assignee
The Fourth Paradigm (Beijing) Tech Co. Ltd.
OA Round
3 (Non-Final)
67%
Grant Probability
Favorable
3-4
OA Rounds
0m
Est. Remaining
99%
With Interview

Examiner Intelligence

Grants 67% — above average
67%
Career Allowance Rate
16 granted / 24 resolved
+11.7% vs TC avg
Strong +62% interview lift
Without
With
+62.5%
Interview Lift
resolved cases with interview
Typical timeline
3y 11m
Avg Prosecution
15 currently pending
Career history
51
Total Applications
across all art units

Statute-Specific Performance

§101
27.2%
-12.8% vs TC avg
§103
50.5%
+10.5% vs TC avg
§102
10.7%
-29.3% vs TC avg
§112
10.7%
-29.3% vs TC avg
Black line = Tech Center average estimate • Based on career data from 24 resolved cases

Office Action

§102 §103
DETAILED ACTION This nonfinal office action is responsive to the amendment filed on June 12, 2026. Claims 1-2, 12-13, 35, and 52-65 are pending. Claim 1 is independent. Claims 3-11 and 14-17 are canceled. Claims 53-65 are added. Claim rejections under 35 USC §102 and §103 have been withdrawn in light of applicant’s amendment. However, new grounds of rejection have been made of claims 1-2, 12-13, 35, and 52-65 under 35 USC §102 and §103 in light of the amendments. See sections Claim Rejections – 35 USC §102 and Claim Rejections – 35 USC §103 below. 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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 June 12, 2026 has been entered. Claim Rejections - 35 USC § 102 In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status. The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action: A person shall be entitled to a patent unless – (a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention. Claim(s) 1, 35, and 52-54 is/are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Murakonda et al. (US11392855), hereinafter Murakonda. Regarding claim 1, Murakonda teaches: providing the GUI; (Murakonda, column 2, lines 28-33: “In example operation , the tool 101 provides a ML design interface that a user can utilize (e.g., via an electronic device such as a personal computer, tablet, or smart phone) to design or configure data pipelines or workflows defining the manner in which ML models are developed, trained, tested, validated, or deployed.” – The ML design interface that a user can utilize is analogous to the GUI.) receiving a dragging operation and a connecting operation of a user in the GUI to generate a training data node and a model training operator node in the GUI, (Murakonda, column 2, lines 34-45: “The tool 101 may display a canvas area and one or more “libraries” of templates that a user can drag and drop into the canvas area. For example, a user may drag and drop into the canvas area an icon representing a particular ML model or algorithm. The user may then drag and drop into the canvas area one or more icons representing datastores. The user may connect the datastore icons to the model icon, and may provide input indicating that the data is to be utilized for developing the model (e.g., when the data is training data, testing data, or validation data) or indicating that the data is to be utilized when the model is deployed (e.g., when the data is input data that will be utilized to develop predictive results).” – The dragging and dropping of the model and dataset, where it is indicated that it is utilized in developing the model, is analogous to the dragging and connecting operation to generate a training data node and a model training operator node.) and establishing a first connection between the training data node and the model training operator node, (Murakonda, column 2, lines 46-48: “Similarly, the user may drag and drop, into the canvas area, a datastore icon and may connect the datastore icon to the model icon.”) wherein the training data node represents training data, and the model training operator node represents a model training operator; (Murakonda, column 5, lines 65-66: “FIG . 2 depicts an example method 200 for developing and utilizing a model that may be implemented via the tool 101.” – where 202 represents the training data node and 204 represents the model training operator node.) training a machine learning model based on the training data using the model training operator; (Murakonda, column 6, lines 37-39: “Regarding the sub - step 242 of step 204 , developing the model typically involves training the model using training data.”) receiving a dragging operation and a connecting operation of the user in the GUI to generate a prediction data node and a model prediction operator node in the GUI, (Murakonda, column 2, lines 34-45: “The tool 101 may display a canvas area and one or more “libraries” of templates that a user can drag and drop into the canvas area. For example, a user may drag and drop into the canvas area an icon representing a particular ML model or algorithm. The user may then drag and drop into the canvas area one or more icons representing datastores. The user may connect the datastore icons to the model icon, and may provide input indicating that the data is to be utilized for developing the model (e.g., when the data is training data, testing data, or validation data) or indicating that the data is to be utilized when the model is deployed (e.g., when the data is input data that will be utilized to develop predictive results).” – The dragging and dropping of the model and dataset, where it is indicated that it is utilized in deploying the model, is analogous to the dragging and connecting operation to generate a data node and a model prediction operator node.) establishing a second connection between the model training operator node and the model prediction operator node, and establishing a third connection between the prediction data node and the model prediction operator node, (Murakonda, column 2, lines 46-48: “Similarly, the user may drag and drop, into the canvas area, a datastore icon and may connect the datastore icon to the model icon.”) wherein the prediction data node represents prediction data, the model prediction operator node represents a model prediction operator, and the model prediction operator and the model training operator are mutually independent; (Murakonda, Figure 2 – Where 262 and 264 represent the prediction data node, 206 represents the model prediction operator. 204, the model training operator, and 206, the model prediction operator, are independent of each other.) wherein the second connection enables a trained machine learning model to be configured for use by the model prediction operator; and (Murakonda, column 14, lines 41-42: “After the model has been trained, the model may be implemented to develop predictive results.”) providing a prediction service on the prediction data using the model prediction operator. (Murakonda, column 11, lines 48-54: “The server 107 and the client 105 coordinate to enable the tool 101 to: (i) initiate a configuration session; (ii) detect a workflow defined by a user; (iii) instantiate the workflow and associated objective; (iv) develop the model included in the workflow; (v) implement the model in the workflow to develop predictive results; and (vi) provide interactive visualizations of the predictive results.” – Where steps (v)-(vi) are providing a prediction service on the prediction data using the model prediction operator.) Regarding claim 35, Murakonda teaches the method of claim 1, as cited above. Murakonda further teaches: An apparatus comprising at least one computing device and at least one storage device, wherein the at least one storage device is configured to store instructions, and wherein the instructions are configured, upon being executed by the at least one computing device, to cause the at least one computing device to execute the method of claim 1 for execution of an automated machine learning process. (Murakonda, column 11, lines 20-27: “The memory 105a of the client 105 includes a client module 106 (sometimes referred to as the “module 106”). The module 106 is a set of machine readable instructions (e.g., a software module, application, or routine) that, when executed by the processor 105b, causes the client 105 to communicate with the server 107 and to receive from server 107 data utilized to provide a graphical user (GUI) for interacting with the pipeline tool 101.”) Regarding claim 52, Murakonda teaches the method of claim 1, as cited above. Murakonda further teaches: A computer-readable storage medium having a computer program stored thereon, which computer program, when executed by a processor, implements the method of claim 1. (Murakonda, column 15, lines 62-67: “When implemented in software , any of the applications, services, and engines described herein may be stored in any tangible, non-transitory computer readable memory such as on a magnetic disk, a laser disk, solid state memory device, molecular memory storage device, or other storage medium, in a RAM or ROM of a computer or processor, etc.”) Regarding claim 53, Murakonda teaches the method of claim 1, as cited above. Murakonda further teaches: providing a configuration interface for configuring model training in response to triggering operation on the model training operator; (Murakonda, column 13, lines 22-24: “To initiate the configuration session, the client 105 may transmit to the server 107 a request to begin the session.” And lines 37-40: “After the client 105 launches the GUI, a user may interact with the GUI to build the workflow. For example, the user may place a set of icons and connections in a “configuration area” of the GUI.” – The request to begin the session is analogous to the triggering operation, where the configuration area is analogous to the configuration interface.) obtaining training samples by data preprocessing and feature engineering processing on the training data according to configuration information input through the configuration interface; and (Murakonda, column 6, lines 12-19: “At the steps 228 and 230, the tool 101 prepares the training data, testing data, or validation data for use by the model. Generally speaking, data preparation ensures that data is consumable by a model in an expected manner. Data preparation may include data classification, data joins, data softening/hardening, data features extraction, data transformation, data aggregation and sorting, etc.” – The data preparation is analogous to the data preprocessing and feature engineering.) training the machine learning model based on the training samples using at least one model training algorithm. (Murakonda, column 7, lines 32-35: “The server 107 may be coupled via one or more links or networks 112 to any one or more of: (i) a set of databases storing templates that can be utilized design or configure a workflow;” and column 8, lines 16-17: “Finally, the model templates 172-176 may each represent a different type of ML model or algorithm.”) Regarding claim 54, Murakonda teaches the method of claim 53, as cited above. Murakonda further teaches: the configuration interface includes at least one of: an input source configuration of the machine learning model; (Murakonda, column 8, lines 39-45: In operation, a user may select a particular one of the databases 121 as a data source for a model by selecting an icon representing one of the templates 160, dragging and dropping the icon into a canvas or configuration area of a GUI, and interacting with the icon or GUI to provide on or more properties (e.g., the IP address) particular to the particular database 121.”) an applicable problem type configuration of the machine learning model; an algorithm mode configuration for training the machine learning model; an optimization objective configuration of the machine learning model; or a field name configuration of a prediction objective field of the machine learning model. (Murakonda, column 7, lines 64-65: “Example information input by a user may include: a database name, table, query, etc.” – The information input by the user is analogous to the field name configuration.) Claim Rejections - 35 USC § 103 The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action: A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made. Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Murakonda in view of Bourhani et al. (US20170316114), hereinafter Bourhani. Regarding claim 2, Murakonda teaches the method of claim 1, as cited above. Murakonda does not explicitly teach: providing an editing interface according to operation of editing the model training operator; acquiring operator content input through the editing interface, wherein the operator content input includes an operation command of data preprocessing on input training data, an operation command of feature engineering on the training data that has undergone data preprocessing, and an operation command of model training according to results of feature engineering; and encapsulating the operator content input to obtain the model training operator. However, Bourhani teaches: providing an editing interface according to operation of editing the model training operator; (Bourhani, paragraph 0026: “Within the Data Preparation category in the node palette 204, a Transformation/Bin node 222, a Feature Selection node 224, a Create New Variable node 226, and a Sample/Balance node 228 are each selectable.” And paragraph 0029: “The node action selector menu 258 may be populated with actions available for the specific nodes (e.g., the nodes in the example model 244) on the digital canvas 206. Various possible actions include an edit action to change a node configuration. The edit action causes the GUI 200 to present information regarding the settings of the node (see FIG. 3) and allow the user to change its configuration thereby changing the node's behavior.” – The digital canvas is analogous to the canvas taught by Murakonda above which is analogous to the model training operator. The edit action is analogous to the editing interface.) acquiring operator content input through the editing interface, wherein the operator content input includes an operation command of data preprocessing on input training data, (Bourhani, paragraph 0026: “The Transformation/Bin node 222 modifies the shapes of distribution of numerical (e.g., continuous) variables either by applying numerical transformations or binning techniques.” – The transformation is one of the actions able to be done in the edit actions taught above and is a command of data preprocessing of the training data.) an operation command of feature engineering on the training data that has undergone data preprocessing, and an operation command of model training according to results of feature engineering; and (Bourhani, paragraph 0026: “The Feature Selection node 224 screens variables by looking for structural problems in the fields as well as taking statistical tests of significance of the dependency between each predictor and the target variable, and allows the user set what variables to keep.”- The feature selection node is one of the actions able to be done in the edit actions taught above and the capability to set which variables to keep is analogous to feature engineering.) encapsulating the operator content input to obtain the model training operator. (Bourhani, paragraph 0021: “The graphical linking pattern establishes that the output of the first node 246 is provided to the input of the second node 248, the output of the second node 248 is provided to the input of the third node 250, and so forth.” – The nodes being connected sequentially means that the operator content input that is being done through the editing interface is sent to the model training node and is therefore encapsulated to obtain the model training operator.) Bourhani is considered analogous to the claimed invention as it is in the same field of endeavor, machine learning and creating automated machine learning pipelines. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to have modified Murakonda, which already teaches the GUI containing the model training operator but does not explicitly teach an editing interface that includes commands to preprocess data and conduct feature engineering, to include the teachings of Bourhani which does teach an editing interface that includes commands to preprocess data and conduct feature engineering in order to present "an analytical model in a visual manner that is user friendly and with which data scientist and engineers can easily interact." (Bourhani, paragraph 0010) Claims 12-13 and 62-65 are rejected under 35 U.S.C. 103 as being unpatentable over Murakonda in view of Bansal et al. (US20210097444), hereinafter Bansal. Regarding claim 12, Murakonda teaches the method of claim 1, as cited above. Murakonda further teaches: obtaining a model training scheme based on the trained machine learning model; and (Murakonda, column 9, lines 8-16: “Generally speaking, each object is instantiated from a template or class having a number of preconfigured settings or properties particular to that template or class. For example, a first template or class may exist for a KNN model and a second template or class may exist for a logistic regression. The first template may include the variables and logic necessary to implement a KNN algorithm and the second template may include the variables and logic necessary to implement a logistic regression algorithm.” – The templates are analogous to the model training scheme.) wherein the model training scheme includes any one or more of: an algorithm used to train the machine learning model, hyperparameters of the machine learning model, effects of the machine learning model, (Murakonda, column 9, lines 39-49: “Regarding model objects such as the model object 148, each model object represents a particular model or algorithm that is or will be utilized in a workflow (e.g., from one of the model templates 113). Each model object may include or reference information particular to an instantiated object, such as a unique ID, a model type, an ID of one or more data source objects, an ID of one or more data destination objects, a mapping of the input and output objects to particular inputs and outputs of the model (e.g., to indicate which of the data source objects is utilized for training data, validation data, testing data, or input data during model implementation) .”) Murakonda does not explicitly teach: visualizing the model training scheme, that the model training scheme includes feature information. wherein the feature information includes any one or more of feature quantity, feature generation method and feature importance analysis results. However, Bansal teaches: visualizing the model training scheme, (Bansal, paragraph 0063: “For additional detail, FIG. 7 is a diagram illustrating one exemplary user interface 700 for viewing trial results of an automated machine learning pipeline exploration according to some embodiments.” – Viewing trial results is analogous to visualizing the model training scheme.) that the model training scheme includes feature information. (Bansal, figure 8 – The “define feature processing pipelines” panel gives feature information.) wherein the feature information includes any one or more of feature quantity, feature generation method and feature importance analysis results. (Bansal, paragraph 0066: “As shown, a first pipeline “FP_BASELINE” is defined with a first “FPO” step, which is defined (as a training job to learn the transformations) with values for a source directory, instance types and counts, and other non-illustrated values such as an ML framework version to be used, a set of tags to be applied, an identifier of a feature processing strategy (e.g., a baseline strategy that performs a 1-hot encoding of all categorical variables and does a median-impute null values with indicators;” – This is describing the feature processing pipelines in figure 8, the feature processing strategy is analogous to the feature generation method.) Bansal is considered analogous to the claimed invention as it is in the same field of endeavor, machine learning and creating automated machine learning pipelines. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to have modified Murakonda, which already teaches training a model and obtaining a model training scheme but does not explicitly teach visualizing the model training scheme which includes feature information of a feature quantity, feature generation method, or feature importance analysis results, to include the teachings of Bansal which does teach visualizing the model training scheme which includes feature information of a feature quantity, feature generation method, or feature importance analysis results in order to make it easier for those who don't have the required knowledge to "clean up the data or preprocess in order to build good models." (Bansal, paragraph 0002) Regarding claim 13, Murakonda and Bansal teach the method of claim 12, as cited above. Murakonda does not explicitly teach: retraining the machine learning model according to preview results of the visualizing. However, Bansal further teaches: retraining the machine learning model according to preview results of the visualizing. (Bansal, paragraph 0073: “900 further include receiving a fourth request message originated by the computing device of the user indicating a request to perform a modified pipeline of one of the plurality of ML pipeline plans, wherein the modified pipeline was specified by the user based on the one ML pipeline plan.”) Regarding claim 62, Murakonda teaches the method of claim 1, as cited above. Murakonda further teaches: obtaining prediction samples by data preprocessing and feature-update processing on the prediction data using the model prediction operator according to configuration information input by the user through the configuration interface; and (Murakonda, column 6, lines 12-19: “At the steps 228 and 230, the tool 101 prepares the training data, testing data, or validation data for use by the model. Generally speaking, data preparation ensures that data is consumable by a model in an expected manner. Data preparation may include data classification, data joins, data softening/hardening, data features extraction, data transformation, data aggregation and sorting, etc.”) providing prediction results for the prediction samples using the trained machine learning model. (Murakonda, “The user may connect the datastore icons to the model icon, and may provide input indicating that the data is to be utilized for developing the model (e.g., when the data is training data, testing data, or validation data) or indicating that the data is to be utilized when the model is deployed (e.g., when the data is input data that will be utilized to develop predictive results).” And column 1, lines 43-45: “Once deployed, the designed ML model may generate predictive results based on input data fed to the ML model.”) Murakonda does not explicitly teach: providing a configuration interface for configuring a batch prediction service in response to a triggering operation of the model prediction operator; However, Bansal teaches: providing a configuration interface for configuring a batch prediction service in response to a triggering operation of the model prediction operator; (Bansal, paragraph 0060: “Thus, as indicated above, at optional circle (7C) the AMPGS 102 may send one or more commands (e.g., API calls) to a batch inference system 142 to “host” the pipeline for use in performing batch inferences. The batch inference system 142 may be implemented similarly to the model hosting system 140 described herein, but may be optimized for performing inferences in batch—e.g., via input data stored at a storage service location, storing results back to a storage location, etc., as opposed to performing real-time synchronous inference.”) Bansal is considered analogous to the claimed invention as it is in the same field of endeavor, machine learning and creating automated machine learning pipelines. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to have modified Murakonda, which already teaches providing a prediction service using a model prediction operator but does not explicitly teach configuring a batch prediction service, to include the teachings of Bansal which does teach configuring a batch prediction service in order to have the capability of providing inference in either real time or batch service. (Bansal, paragraph 0018) Regarding claim 63, Murakonda and Bansal teach the method of claim 62, as cited above. Murakonda does not explicitly teach: a configuration of field selection in a prediction result, or a configuration of switching state of a simulated real-time prediction service. However, Bansal further teaches: a configuration of field selection in a prediction result, or (Bansal, paragraph 0030: “This and other user interfaces are shown as including various exemplary data fields and user interface input elements (e.g., text input boxes, drop down boxes, etc.)”) a configuration of switching state of a simulated real-time prediction service. Regarding claim 64, Murakonda and Bansal teach the method of claim 62, as cited above. Murakonda does not explicitly teach: providing a configuration interface for configuring a real-time prediction service according to an operation of configuring the real-time prediction service; receiving a prediction service request including the prediction data through an API address provided in the configuration interface; and obtaining a prediction result on the prediction data in response to the received prediction service request using the machine learning model selected, and sending the prediction result through the API address. However, Bansal further teaches: providing a configuration interface for configuring a real-time prediction service according to an operation of configuring the real-time prediction service; (Bansal, paragraph 0059: “As one example, the commands may include a “create model” API call that combines code for the model (e.g., inference code implemented within a container) along with model artifacts (e.g., data describing weights associated with various aspects of the model) to yield a model that can be hosted (e.g., by a model hosting system 140 or batch inference system 142)”) receiving a prediction service request including the prediction data through an API address provided in the configuration interface; and (Bansal, paragraph 0029: “The user 109 may, at circle (2), utilize the computing device 104 to initiate ML pipeline exploration by the AMPGS 102. For example, the user 109 may cause the computing device 104 to issue one or more requests (e.g., API calls) to the ML service 110, which may occur through the user 109 configuring an ML exploration job using one or more user interfaces such as those depicted in FIGS. 2-4.” And paragraph 0059: “As one example, the commands may include a “create model” API call that combines code for the model (e.g., inference code implemented within a container) along with model artifacts (e.g., data describing weights associated with various aspects of the model) to yield a model that can be hosted (e.g., by a model hosting system 140 or batch inference system 142),” – The API call to initiate the ML pipeline is analogous to the prediction service request through an API address.) obtaining a prediction result on the prediction data in response to the received prediction service request using the machine learning model selected, and sending the prediction result through the API address. (Bansal, paragraph 0029: “Such user interfaces may be provided by the ML service 110 (or by another service of the provider network), and the user 109 may provide user input via these interfaces which is ultimately provided to the AMPGS 102 to configure, launch, and/or modify an ML exploration job.” – After the API call, the results are sent as part of the ML exploration job.) Regarding claim 65, Murakonda and Bansal teach the method of claim 64, as cited above. Murakonda does not explicitly teach: a configuration for model selection rules for selecting an online machine learning model from the trained machine learning models, or a configuration for application resources. However, Bansal further teaches: a configuration for model selection rules for selecting an online machine learning model from the trained machine learning models, or (Bansal, paragraph 0036: “In this user interface 400, a user may specify via input element 405 what machine learning problem type is at hand, enabling the use to control what type of ML algorithms will be used for training. For example, many different machine learning problem types are known to those of skill in the art, such as binary classification, multi-class classification, linear regression, and so on. In some embodiments, the user may be able to select an “auto detection” option, as the machine learning problem type may be inferred based on the type and/or values of the “target column” the user specified.” – The user is able to select “auto detection” which infers the best algorithm based on the type and/or values which is analogous to selection rules for selecting the model.) a configuration for application resources. (Bansal, paragraph 0034: “In this example, a user can also use user interface 300 input elements 315/320 to indicate what type of computing instances (e.g., virtual machine types, where multiple different options may be made available having different resource characteristics, such as differing amounts of memory, processing capability, network availability, storage, etc.) are to be used for the ML model training jobs (via element 315), and what number of those instances are to be used (via element 320) at any point in time, allowing the user to control how fast the exploration occurs and at what expense.”) Claims 55 and 56 are rejected under 35 U.S.C. 103 as being unpatentable over Murakonda in view of Wahner (Data Preprocessing vs. Data Wrangling in Machine Learning Projects), hereinafter Wahner. Regarding claim 55, Murakonda teaches the method of claim 54, as cited above. Murakonda does not explicitly teach: the data preprocessing on the training data comprises at least one of: performing data type conversion of the training data; sampling the training data; annotating the training data as labeled data and unlabeled data; automatically identifying a data field type of the training data; filling in missing values of the training data; analyzing an initial time field of the training data, obtaining and adding a new time field, and deleting the initial time field; or automatically identifying non-numerical data in the training data, and hashing the non- numerical data. However, Wahner teaches: the data preprocessing on the training data comprises at least one of: performing data type conversion of the training data; sampling the training data; annotating the training data as labeled data and unlabeled data; automatically identifying a data field type of the training data; filling in missing values of the training data; analyzing an initial time field of the training data, obtaining and adding a new time field, and deleting the initial time field; or automatically identifying non-numerical data in the training data, and hashing the non- numerical data. (Wahner, page 2, paragraph 2: “Data Cleansing puts data into the right shape and quality for analysis. It includes many different functions, for example the following: Basics (select, filter, removal of duplicates, ... ) Sampling (balanced, stratified, ... ) Data Partitioning (create training+ validation+ test data set, ... ) Transformations (normalisation, standardisation, scaling, pivoting, ... ) Binning (count-based, handling of missing values as its own group, ... ) Data Replacement (cutting, splitting, merging, ... ) Weighting and Selection (attribute weighting, automatic optimization, ... ) Attribute Generation (ID generation, ... ) Imputation (replacement of missing observations by using statistical algorithms) – Murakonda teaches data preparation, which is analogous to the data preprocessing, as cited above. Wahner’s data preparation is analogous to the data preprocessing and includes transformation (data conversion), sampling, and imputation (filling in missing values)) Wahner is considered analogous to the claimed invention as it is in the same field of endeavor, machine learning. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to have modified Murakonda, which already teaches data preprocessing but does not explicitly teach the different subtypes of preprocessing specifically mentioned above, to include the teachings of Wahner which does teach a number of different ways to preprocess the data as this is a “fundamental application of machine learning and deep learning.” (Wahner, page 4, paragraph 2) Regarding claim 56, Murakonda teaches the method of claim 53, as cited above. Murakonda does not explicitly teach: sampling the training data that has undergone the data preprocessing; performing feature pre-selection on the training data that has undergone the sampling, to obtain basic features; performing feature derivation on the basic features to obtain derived features; and generating the training samples according to the basic features and the derived features. However, Wahner teaches: sampling the training data that has undergone the data preprocessing; (Wahner, page 2, paragraph 2: “Sampling (balanced, stratified, ... )”) performing feature pre-selection on the training data that has undergone the sampling, to obtain basic features; (Wahner, page 7, paragraph 1: “Creation of a new column via feature extraction: Get the name prefix of each passenger out of its long name to defer its sex, e.g. Mr, Mrs, Miss, Master” – The feature extraction is analogous to the feature pre-selection to obtain basic features.) performing feature derivation on the basic features to obtain derived features; and (Wahner, page 7, paragraph 1: “Creation of a new column via aggregation to see how many people were in the traveling group of each passenger: "family size = 1 + SibSp + Parch"” – Creating new “columns” of features is analogous to feature derivation and it uses the basic features.) generating the training samples according to the basic features and the derived features. (Wahner, page 7, paragraph 1: “Therefore, the initial dataset needs to be prepared well to produce best results when applying a machine learning algorithm.” – The preparation above is the new training samples that include the basic features and derived features.) Wahner is considered analogous to the claimed invention as it is in the same field of endeavor, machine learning. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to have modified Murakonda, which already teaches feature engineering but does not explicitly teach feature preselection, feature derivation and generating training samples from the basic and derived features, to include the teachings of Wahner which does teach feature preselection, feature derivation and generating training samples from the basic and derived features as this is a “fundamental application of machine learning and deep learning.” (Wahner, page 4, paragraph 2) Claims 57 and 59 are rejected under 35 U.S.C. 103 as being unpatentable over Murakonda in view of Wahner in view of Bansal. Regarding claim 57, Murakonda and Wahner teach the method of claim 56, as cited above. Murakonda does not explicitly teach: extracting all attribute information included in the training data that has undergone the sampling, wherein the attribute information is used to form features; acquiring feature importance values of each attribute information; and obtaining the basic features according to the feature importance values. However, Wahner further teaches: extracting all attribute information included in the training data that has undergone the sampling, wherein the attribute information is used to form features; (Wahner, page 2, paragraph 2: “Attribute Generation (ID generation, ... )” and paragraph 3: “Feature Engineering selects the right attributes to analyze.” – The attribute generation is analogous to extracting all attribute information.) Murakonda and Wahner do not explicitly teach: acquiring feature importance values of each attribute information; and obtaining the basic features according to the feature importance values. However, Bansal teaches: acquiring feature importance values of each attribute information; and (Bansal, paragraph 0044: “The pipeline recommender system 112, in some embodiments, can analyze the user's provided dataset and infer one or more of the probabilistic schema of the data set, target leakage, feature importance, the type of ML problem (e.g., classification, regression, etc.) based on the user-identified target column, etc.” – The pipeline recommender system inferring the feature importance of the dataset is analogous to acquiring feature importance values of each attribute information.) obtaining the basic features according to the feature importance values. (Bansal, paragraph 0044: “The pipeline recommender system 112 can use the dataset provided by the user and knowledge learned from metadata collected to recommend a promising and diverse set of feature processing pipelines to apply to the customer dataset along with the code for the feature processing model.” – The feature processing applied to the dataset is analogous to obtaining the basic features according to the feature importance values as that is part of the metadata collected and used to create the features.) Bansal is considered analogous to the claimed invention as it is in the same field of endeavor, machine learning and creating automated machine learning pipelines. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to have modified Murakonda and Wahner, which already teaches feature pre-selection but does not explicitly teach acquiring feature importance values and obtaining basic features according to the feature importance values, to include the teachings of Bansal which does teach acquiring feature importance values and obtaining basic features according to the feature importance values in order to allow “users to have visibility into the details of the resulting ML model, the details of the training job leading to the ML model, the parameters and operations of the ML pipeline, and optionally some or all of these details of other ML pipelines that were explored yet ultimately determined to be less-performant.” (Bansal, paragraph 0018) Regarding claim 59, Murakonda and Wahner teach the method of claim 56, as cited above. Murakonda and Wahner do not explicitly teach: performing at least one of statistical calculation or feature combination on the basic features to obtain the derived features, using preset feature generation rules. However, Bansal teaches: performing at least one of statistical calculation or feature combination on the basic features to obtain the derived features, using preset feature generation rules. (Bansal, paragraph 0050: “As one example, the pipeline recommender system 112 may recommend up to ten pipelines to explore, such as (1) “apply one-hot encoding and principal component analysis (as the feature preprocessors/transforms) followed by use of the ‘XGBOOST’ algorithm with hyperparameter tuning”, (2) “apply one-hot encoding and principal component analysis (as the feature preprocessors/transforms) followed by use of a ‘linear learner’ algorithm with hyperparameter tuning”, (3) “apply principal component analysis (as the feature preprocessor/transform) followed by use of the ‘XGBOOST’ algorithm with hyperparameter tuning”, and the like.” – one-hot encoding, principal component analysis, and XGBOOST are all examples of feature generation rules. One-hot encoding and principal component analysis are both methods of feature combination while XGBOOST incorporates the L1 or L2 norm and thus involves a statistical calculation.) Bansal is considered analogous to the claimed invention as it is in the same field of endeavor, machine learning and creating automated machine learning pipelines. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to have modified Murakonda and Wahner, which already teaches obtaining derived features but does not explicitly teach that obtaining derived features is done with a statistical calculation or feature combination on the basic features using preset feature generation rules, to include the teachings of Bansal which does teach obtaining derived features is done with a statistical calculation or feature combination on the basic features using preset feature generation rules in order to allow “users to have visibility into the details of the resulting ML model, the details of the training job leading to the ML model, the parameters and operations of the ML pipeline, and optionally some or all of these details of other ML pipelines that were explored yet ultimately determined to be less-performant.” (Bansal, paragraph 0018) Claims 58, 60, and 61 are rejected under 35 U.S.C. 103 as being unpatentable over Murakonda in view of Wahner in view of Bansal in view of Ved (Feature Selection in Machine Learning: Variable Ranking and Feature Subset Selection Methods), hereinafter Ved. Regarding claim 58, Murakonda, Wahner, and Bansal teach the method of claim 57, as cited above. Murakonda, Wahner, and Bansal do not explicitly teach: ranking all the feature importance values to obtain a ranking result; and acquiring a first predetermined quantity of the attribute information as the basic features according to the ranking result. However, Ved teaches: ranking all the feature importance values to obtain a ranking result; and (Ved, Page 2, last 2 paragraphs: “Variable Ranking is the process of ordering the features by the value of some scoring function, which usually measures feature-relevance. Resulting set: The score S(fi) is computed from the training data, measuring some criteria of feature fi. By convention a high score is indicative for a valuable (relevant) feature.” – The basic features obtained by Murakonda, Wahner, and Bansal in claim 57 are analogous to the training data which is then used to score (rank) the features. These scores are analogous to the ranking results.) acquiring a first predetermined quantity of the attribute information as the basic features according to the ranking result. (Ved, page 3, paragraph 1: “A simple method for feature selection using variable ranking is to select the k highest ranked features according to S. This is usually not optimal, but often preferable to other, more complicated methods. It is computationally efficient — only calculation and sorting of n scores.” – The k highest ranked features is indicative that k is a predetermined quantity.) Ved is considered analogous to the claimed invention as it is in the same field of endeavor, machine learning. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to have modified Murakonda, Wahner, and Bansal, which already teaches obtaining basic features according to feature importance but does not explicitly teach ranking the feature importance values to obtain a ranking result and acquiring a predetermined number of attributes based on that, to include the teachings of Ved which does teach ranking the feature importance values to obtain a ranking result and acquiring a predetermined number of attributes based on that in order to "select variables according to their individual predictive power." (Ved, page 6, last paragraph) Regarding claim 60, Murakonda, Wahner, and Bansal teach the method of claim 59, as cited above. Murakonda, Wahner, and Bansal do not explicitly teach: performing feature post-selection on the basic features and the derived features; and generating training samples according to features obtained through the feature post- selection. However, Ved teaches: performing feature post-selection on the basic features and the derived features; and (Ved, page 7, paragraph 2: “For Feature Subset Selection you’d need: A measure for assessing the goodness of a feature subset (scoring function)” – The specification states that feature post-selection includes “acquiring feature importance values of each basic feature and each derived feature” in paragraph 00104. Therefore, feature subset selection which includes a scoring function is analogous to feature post-selection, given the set of basic features and derived features from claim 53.) generating training samples according to features obtained through the feature post- selection. (Ved, page 7, paragraph 1: “The Goal of Feature Subset Selection is to find the optimal feature subset.” – The optimal feature subset is analogous to the training samples as the subset is being generated based off the feature post-selection.) Ved is considered analogous to the claimed invention as it is in the same field of endeavor, machine learning. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to have modified Murakonda, Wahner, and Bansal, which already teaches obtaining derived features but does not explicitly teach performing post-selection on the basic and derived features and generating training samples according to that, to include the teachings of Ved which does teach performing post-selection on the basic and derived features and generating training samples since it provides “generic selection of features, not tuned by given learner (universal).” (Ved, page 8, fourth bullet point. Regarding claim 61, Murakonda, Wahner, and Bansal teach the method of claim 60, as cited above. Murakonda, Wahner, and Bansal do not explicitly teach: acquiring feature importance values of each basic feature and each derived feature, ranking all the feature importance values to obtain a ranking result; and acquiring a second predetermined quantity of features as required features for generating the training samples, according to the ranking result. However, Ved teaches: acquiring feature importance values of each basic feature and each derived feature, ranking all the feature importance values to obtain a ranking result; and (Ved, Page 2, last 2 paragraphs: “Variable Ranking is the process of ordering the features by the value of some scoring function, which usually measures feature-relevance. Resulting set: The score S(fi) is computed from the training data, measuring some criteria of feature fi. By convention a high score is indicative for a valuable (relevant) feature.” – The basic features obtained by Murakonda, Wahner, and Bansal in claim 57 are analogous to the training data which is then used to score (rank) the features. These scores are analogous to the ranking results.) acquiring a second predetermined quantity of features as required features for generating the training samples, according to the ranking result. (Ved, page 3, paragraph 1: “A simple method for feature selection using variable ranking is to select the k highest ranked features according to S. This is usually not optimal, but often preferable to other, more complicated methods. It is computationally efficient — only calculation and sorting of n scores.” – The k highest ranked features is indicative that k is a predetermined quantity.) Response to Arguments Applicant’s arguments with respect to claim(s) 1 have been considered but 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. Conclusion The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. McClure (GUI-fying the Machine Learning Workflow: Towards Rapid Discovery of Viable Pipelines) discusses a GUI for a machine learning workflow. Nookula et al. (US11853401) discusses a model training and deployment using building blocks via a GUI. Any inquiry concerning this communication or earlier communications from the examiner should be directed to JACQUELINE MEYER whose telephone number is (703)756-5676. The examiner can normally be reached M-F 8:00 am - 4:00 pm EST. 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, Tamara Kyle can be reached at 571-272-4241. 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. /J.C.M./Examiner, Art Unit 2144 /TAMARA T KYLE/Supervisory Patent Examiner, Art Unit 2144
Read full office action

Prosecution Timeline

Oct 17, 2022
Application Filed
Sep 11, 2025
Non-Final Rejection mailed — §102, §103
Dec 11, 2025
Response Filed
Mar 12, 2026
Final Rejection mailed — §102, §103
Jun 12, 2026
Request for Continued Examination
Jun 17, 2026
Response after Non-Final Action
Oct 30, 2026
Non-Final Rejection mailed — §102, §103 (current)

Precedent Cases

Applications granted by this same examiner with similar technology

Patent 12768877
METHOD FOR ESTIMATING PERFORMANCE VALUES OF CHIPS, COMPUTING SYSTEM, AND NON-TRANSITORY COMPUTER-READABLE STORAGE MEDIUM
2y 10m to grant Granted Oct 20, 2026
Patent 12743602
Automated Variational Inference using Stochastic Models with Irregular Beliefs
3y 8m to grant Granted Sep 22, 2026
Patent 12737637
MACHINE LEARNING MODEL TRAINING WITH ADVERSARIAL LEARNING AND TRIPLET LOSS REGULARIZATION
3y 6m to grant Granted Sep 15, 2026
Patent 12717933
METHOD AND SYSTEM FOR SECURING NEURAL NETWORK MODELS
4y 2m to grant Granted Aug 25, 2026
Patent 12718094
SYSTEM AND METHOD FOR CONTINUAL REFINABLE NETWORK
3y 8m to grant Granted Aug 25, 2026
Study what changed to get past this examiner. Based on 5 most recent grants.

Strategy Recommendation AI-generated — please review before filing

Get a prosecution strategy drawn from examiner precedents, rejection analysis, and claim mapping.
Typically takes 5-10 seconds — AI-generated, attorney review required before filing

Prosecution Projections

3-4
Expected OA Rounds
67%
Grant Probability
99%
With Interview (+62.5%)
3y 11m (~0m remaining)
Median Time to Grant
High
PTA Risk
Based on 24 resolved cases by this examiner. Grant probability derived from career allowance rate.

Sign in with your work email

Enter your email to receive a magic link. No password needed.

Personal email addresses (Gmail, Yahoo, etc.) are not accepted.

Free tier: 3 strategy analyses per month