DETAILED ACTION
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 .
Status of Claims
The following is a final office action.
Claims [1, 4, 6, 8-11, 14-15, 18 and 20-29] are currently pending and have been examined.
Claims 5, 7, 12, 16-17, and 19 are newly canceled see REMARKS January 22, 2026
Claims 1, 8, 11, 14, 18, and 20-21 are amended see REMARKS January 22, 2026.
Claims 22-29 are newly added see REMARKS January 22, 2026.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1, 4, 6, 8-11, 14-15, 18 and 20-29 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception that is an abstract idea without a practical application or significantly more.
Step 1: Claims 1, 4, 6, 8-10, and 27-28 recite a system, Claims 11, 14-15, 18, 20, and 29 recite a method (i.e. a process such as an act or series of steps) and claims 21-26 recites a computer program product comprising a non-transient storage device and therefore each claim falls within one of the four statutory categories.
Step 2A prong 1 (Is a judicial exception recited?):
The representative claim 1 recites: detecting anomalies in account data, comprising: receive unlabeled account data sets comprising data points with corresponding feature values for defined input features as training data, clustering the account data sets into a set of clusters based on similarities between the feature values for the input features within each cluster being more than across other clusters; receive the set of clusters and corresponding account data sets contained within each of the clusters; determine, for each of the clusters, a distribution pattern of the feature values in the account data sets, corresponding to a plurality of accounts, for a particular feature defined as being associated with detecting anomalies and based on the distribution pattern, determine a percentile threshold value above which anomalies occur for the particular feature and label the data points in each of the account data sets for each cluster having the feature values for the particular feature exceeding the percentile threshold value with anomaly metadata indicative of anomaly and others as normal to generate labelled data sets with the anomaly metadata; wherein labelling the data points comprises identifying particular data points being outliers in each cluster which includes determining, from the distribution pattern for each of the clusters, a deviation amount from a median of the distribution pattern which corresponds to a defined percentile occurrence of the particular feature for the account data sets, determining that particular data points having a degree of deviation exceeding the deviation amount thereby indicating anomaly as compared to other data points within that cluster; and receiving the labelled data sets, and mapping the feature values for the input features in the account data sets and extracting a set of rules for generating, dynamically, a rules executable for subsequent real-time classification of anomaly, the rules comprising a set of different combinations of identified features from the input features and corresponding value ranges associated with a likelihood of anomaly for the particular feature, classify new customer data having the input features and apply the set of rules to the feature values of the new customer data to determine a classification of whether the new customer data is outlier or normal.
Claims 11 and 21: A method of anomaly detection in a set of accounts, the method comprising: clustering training data comprising account information into a set of clusters, based on input features for the accounts by: receiving the training data comprising data points defining each feature of the input features for each account in the set of accounts held by an entity, the training data comprising historical data characterizing each said account in terms of the input features for the accounts, each cluster clustering similar accounts having similarities between one or more associated features in the data points; determining, for each of the clusters, a particular feature distribution pattern for accounts contained therein including a median and a degree of deviation, the particular feature defined as related to the anomaly detection; identifying particular data points within each cluster having outlier data based on the particular feature distribution for that cluster and labelling each data point within each cluster as to whether outlier or normal and forming an updated training data set comprising the labelling, wherein identifying the particular data points being outliers in each cluster comprises determining, from the distribution pattern for each of the clusters, a deviation amount from a median of the distribution pattern which corresponds to a defined percentile occurrence of the particular feature for the accounts, and determining that the particular data points in that cluster have a particular degree of deviation exceeding the deviation amount thereby indicative of anomaly as compared to other data points within that cluster; extracting rules to generate, dynamically, a rules executable for anomaly spotting; and, applying the rules executable to new customer data having said feature characteristics to determine a classification in real-time of whether outlier or normal.
The claims recite a mental process. The examiner finds the claims to merely recite a method for determining the likelihood of an anomaly in account data by receiving data, determining a distribution pattern of feature values in the data set, analyzing the data to generate clusters, determining a percentile threshold value above which anomalies occur for a particular feature in each of the account data sets, and generating rules executable for subsequent classification of anomalies. The claims are found to merely recite a series of steps that can be performed in the human mind or with the use of a simple tool such as pen and paper. As the claim limitations recite steps to receive and analyze account data to determine anomalies in the data and using the analysis to determine rules to be used for future analysis can be performed in the human mind by an individual such as a data analyst. Merely receiving data and analyzing the data to determine anomalies is found to be similar to concepts the courts have defined as a mental process including observations, evaluations, judgements, and opinions.
Therefore, the claims are found to recite an abstract idea.
Step 2A Prong 2 (Is the exception integrated into a practical application?): The claims additionally recite additional elements, including;
Claim 1: A computerized machine learning system; an unsupervised clustering module; an anomaly detection module coupled to the unsupervised clustering module; a single tree classification model coupled to the anomaly detection module, sending the classification to a graphical user interface for display.
Claim 11: A computerized method of using machine learning models; a clustering model; training a tree classification model based on the updated training data set being labelled for detecting anomaly; the tree classification model being trained to define combinations of feature characteristics resulting in outlier data, and sending the classification to a graphical user interface for display.
Claim 21: A computer program product comprising a non-transient storage device storing instructions that when executed by at least one processor of a computing device, configure the computing device for using machine learning models; a clustering model; training a tree classification model based on the updated training data set being labelled for detecting anomaly; the tree classification model being trained to define combinations of feature characteristics resulting in outlier data, and sending the classification to a graphical user interface for display.
The additional elements of a computer and a series of models trained to perform data analysis are directed to merely reciting instructions to apply a generic computer and technology to execute the method in the recited claim limitations.
The claim limitations recite mere instructions to implement the abstract idea of detecting anomalies in account data by receiving data and using generic machine learning models to analyze the data.
Therefore, the limitations merely amount to adding the words “apply it” (or an equivalent) to the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea, as discussed in MPEP 2106.05(f) and generally linking the use of the judicial exception to a particular technological environment or field of use, as discussed in MPEP 2106.05(h). Additionally, the additional elements of a generic scanning device being used to scan a product and updating a supply chain distributed ledger based on the event of scanning a product are directed to merely “apply it.” As the claims merely recite a generic scanning device for performing the function of scanning a product to generate an event record to be recorded in a record that is used in the abstract idea.
Furthermore, a method for transmitting, receiving, and processing information does not amount to improvements to the functioning of a computer, or to any other technology or technical field, as discussed in MPEP 2106.05(a), applying the judicial exception with, or by use of, a particular machine, as discussed in MPEP 2106.05(b), effecting a transformation or reduction of a particular article to a different state or thing, as discussed in MPEP 2106.05(c).
Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. As the claims are merely directed to utilizing a computer to perform the steps of receiving data and detecting anomalies by performing generic operations of clustering and processing the data, which are not significant improvements to the functionality of a generic computer, the additional elements are directed to merely “apply it” or applying the abstract idea on a computer.
Step 2B (Does the claim recite additional elements that amount to significantly more than the judicial exception?):
As discussed above, the additional imitations amount to adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, and merely uses a computer as a tool to perform an abstract idea, as discussed in MPEP 2106.05(f). The additional elements of a system comprising generic computer elements and a clustering and tree classification model are not directed to an improvement in a technology or technical field but are merely used to perform the abstract idea of performing a series of steps to detect anomalies in sets of account data. Therefore, the additional elements do not amount to significantly more than the judicial exception.
The dependent claims 4, 6, 8-10, 14-15, 18, 20, and 22-29 further narrow the abstract idea of analyzing account data to detect anomalies as recited in the independent claims 1, 11, and 21.
The dependent claims recite the following additional elements:
Claims 4, 15, and 23: a light gradient boosted model.
Claim 22: wherein the tree classification model is a supervised model and the clustering model is an unsupervised model.
Claim 28: Density-based spatial clustering of applications with Noise
However, the additional elements are directed to merely “apply it” or applying a generic model technique to perform the abstract idea of receiving and analyzing clusters of data to detect potential anomalies.
Therefore, claims 1, 4, 6, 8-11, 14-15, 18 and 20-29 are rejected under 35 U.S.C. 101.
Response to arguments:
Applicant’s arguments, see REMARKS January 22, 2026, and with respect to the rejections of claims [1, 4, 6, 8-11, 14-15, 18 and 20-29] under U.S.C. 101 have been fully considered and are not persuasive.
The representative argues that the independent claims 1, 11, and 21 are directed to a practical application as they recite an improvement in the field of machine learning models dealing with large unlabeled data sets by combining unsupervised and supervised machine learning models as the claims recite steps for an unsupervised model to receive data and generate labels for the data to be used as training data for a supervised model to generate rules for classifying customer data. The representative further argues that the claims do not recite an abstract idea as they cannot be practically performed in the human mind. However, the examiner respectfully disagrees as the claims recite a method for generating rules for classification of anomalies of customer data having input features. The claims recite a method of receiving unlabeled account data sets, clustering the data sets into clusters based on similarities between feature values for the input features, determine a distribution pattern of the feature values in the account data set for a particular feature defined as being associated with detecting anomalies, based on the distribution pattern determine a percentile threshold value above which anomalies occur for a particular feature and label the data points in each of the account data sets for each cluster having the feature values exceeding the threshold and labeling the clusters with a binary value to generate labeled data sets with anomaly metadata, mapping the feature values for the input features in the account data sets and extracting a set of rules comprising a set of different combinations of identified features from the input features and corresponding value ranges associated with a likelihood of anomaly for the particular features, and classifying new customer data having input features and applying the set of rules to determine if a classification of whether the new customer data is outliers or normal. The claims merely recite a method for analyzing customer data to determine patterns indicative of anomalies. A person such as a data analysts would be capable of, mentally or with simple tools such as pen and paper, performing the calculations necessary to identify patterns in customer data indicative of anomalies. Such as receiving unlabeled customer data, clustering the data based on similar features values, determining a distribution pattern, determining a threshold value indicative of anomalies, labeling the anomaly metadata with a binary value indicative of being an anomaly or normal, and determining a set of general rules of determining features indicative of being an anomaly to be used in reviewing new customer data. The claims merely recite concepts the courts have identified as being mental processes such as observation, evaluation, judgment, and opinions, as the claims recite a series of steps for receiving and observing customer data, evaluating the data to determine labels for the presence of anomalies, and generating rules to judge and evaluate subsequent data. Additionally, the examiner finds that the additional elements of a machine learning system comprising an unsupervised clustering module for clustering data sets, an anomaly detection module coupled to the unsupervised clustering module for labeling data points, a single tree classification model for receiving label data sets, generating rules, and classifying new customer data, and a user interface for displaying information are directed to merely “apply it.” As the additional elements merely recite utilizing generic machine learning elements to automate and perform the abstract idea of receiving and analyzing customer data for anomaly detection. Merely using an unsupervised machine learning model to receive and automate the process the steps of labeling information that can subsequently be used to train a supervised machine learning model is not an improvement to a technical field but merely applying a generic technology to perform a generic function. While merely applying a supervised machine learning model to detect anomalies based on a series of rules is not an improvement to the technology of a machine learning model but merely applying a generic model to perform the abstract idea of receiving and analyzing customer information.
The examiner maintains the current 101 rejection.
The applicant argues that the dependent claims 4, 6, 8-10, 14-15, 18, 20, and 22-29 are allowable as being dependent on claims 1, 11, and 21 and therefore are rejected under the same 101 rejection.
Applicant’s arguments, see REMARKS, filed August 18, 2025, with respect to the rejections of 1, 5-6, 10-12, 14, 18, and 20-21 is/are rejected under 35 U.S.C. 103 as being unpatentable over Dodson (US 2018/0316707) in view of Pati (US 2021/0383407) are considered and are persuasive.
Applicant argues that the current prior art does not disclose the newly amended claim limitation “wherein labelling the data points comprises identifying particular data points being outliers in each cluster which includes determining, from the distribution pattern for each of the clusters, a deviation amount from a median of the distribution pattern which corresponds to a defined percentile occurrence of the particular feature for the account data sets, determining that particular data points having a degree of deviation exceeding the deviation amount thereby indicating anomaly as compared to other data points within that cluster” in independent claims 1, 11, and 21. After further search and consideration the examiner agrees that the current prior art individual and in combination do not disclose the newly amended claim limitation.
The closes prior art Dodson (US 2018/0316707) discloses a system of analyzing received data by using an unsupervised machine learning model to receive and cluster data sets, labeling data sets by applying a threshold for determining anomalies, and training a rules-based classifier using the labeled data instances (Dodson [0003-0004]). Dodson further discloses clustering data sets and generating vectors to subsequently label data sets as outliers or inlier (1, 0) by applying a threshold to the overall measure of an outlier-ness of the data and generating a training set for a classifier (Dodson [0143]).
The second closes prior art Pati (US 2021/0383407) teaches the training and fine tuning of an anomaly detection model (Pati [0015]). Pati further teaches a system of using an unsupervised machine learning model such as an isolation forest model to determine anomalous patterns form a data set by providing a risk score for each transaction or data set from -1 anomalous to 1 non-anomalous. (Pati [0105-0106]). Once sufficient fraud data is generated and labeled it can be used to train and tune the anomaly detection model. The trained model is then used to perform anomaly detection on data. Pati then sends the identified anomalies to an alert system for presenting to a user via an interface (Pati [0116]).
The third closest prior art Vasseur (US 2019/0138938) teaches a system for training a classifier used to detect network anomalies. Vasseur teaches using a decision tree when analyzing input data to determine anomalies in the data. As well as using a gradient boosting application.
The fourth closest prior art is Serio (US 2014/0025548) which teaches a system of detecting anomalies in transaction information. Serio further teaches receiving and analyzing input information including mortgage and real estate transaction information, debt history, and other account information related to a user.
However, the combination of prior art does not disclose the newly amended claim limitations.
Therefore, the examiner finds claims 1, 11, and 21 allowable over U.S.C. 103.
Therefore, dependent claims 4, 6, 8-10, 14-15, 18, 20, and 22-29 are allowable as being dependent on claims 1, 11, and 21.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure.
Schleith (US 2023/0195715) Systems and methods for detection and correction of anomalies priority.
Baran Pouyan (US 2021/0264306) Utilizing machine learning to detect single and cluster-type anomalies in a dataset.
Dutta (US 2023/0186075) Anomaly detection with model hyperparameter selection.
Givental (US 2021/0281592) Hybrid machine learning to detect anomalies.
Otto (US 2022/0044133) Detection of anomalous data using machine learning.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to COREY RUSS whose telephone number is (571)270-5902. The examiner can normally be reached on M-F 7:30-4:30.
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, Lynda Jasmin can be reached on 5712726782. 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.
/COREY RUSS/Primary Examiner, Art Unit 3629
Prosecution Insights