DETAILED ACTION
This office action is in response to applicant’s response filed on September 9, 2025 in application 18/498,586.
Claims 1-20 are presented for examination. Claims 1-14 are amended. Claims 15-20 are newly added.
IDS submitted on November 22, 2023 was acknowledged.
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 .
Response to Arguments
Applicant’s arguments with respect to claim(s) 1-14 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.
Claim Objections
Claim 6 is objected to because of the following informalities: Claim 6 recited on lines 5-6, “wherein the feature vectors,” as an incomplete statement. Appropriate correction is required.
Claim Rejections - 35 USC § 112
Claims 17 and 20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claims 17 and 20 recited “wherein the internal similarity condition is the maximal distance value between the partial time-series.” Applicant’s specification does not define the internal similarity condition as the maximal distance value. However, the claimed limitation does recite “wherein a cluster of the set of clusters is defined by values of the partial training time-series and the internal similarity condition is a maximal distance value. Examiner interpreted the claimed language to read
partial training time-series + the internal similarity condition = a maximal distance value
Therefore claims 17 and 20 is indefinite without support in the specification. Correction is advised.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1, 3-5, 11, 13, 15-16, 18-19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ye et al. (US 2022/0382622) in further view of Arora et al. (US 2024/0073113).
In regard to claim 1, Ye et al. teach an apparatus for anomaly detection, the apparatus comprising at least one processor and at least one memory including computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to:
collect a measurement time-series relating to a performance indicator (the data store may store any number of tables at any point in time, the tables may be time-series point data values, para. 29);
compute a representative value of said measurement time-series, wherein the representative value is a median value of said measurement time-series or each of said measurement time-series (the centroid may represent an expected or target value for the point data values such as a mean or median value of the historical point data values, para. 37);
provide a clustering model comprising a set of clusters, wherein the clustering model has been trained on a plurality of training time-series relating to the performance indicator (K-means is a clustering algorithm that divides given data points into several clustered centered on centroids, a model trainer generates one or more K-means models based on the historical point data values, para. 37), wherein a cluster of the set of clusters comprises partial time-series that meet an internal similarity condition (the historical point data values 152H may represent a selected or random subset or sampling of the point data values 152 so that the model 212 is trained using less than all of the point data values 152 that the point data anomaly detector 160 receives, para. 33), wherein the cluster of the set of clusters is defined by values of the partial time-series and the internal similarity condition is a maximal distance value, wherein the partial time-series are portions of the plurality of training time-series, wherein the clustering model takes as input the measurement time-series (based on the determined variance value for a given input point data value, determines whether the input point data value is an anomalous point data value, para. 44-48);
select a cluster subset within the set of clusters, wherein the cluster subset is associated with the measurement time-series, wherein the cluster subset comprises at least one cluster which meets an external similarity condition with the measurement time-series, wherein the external similarity condition is a function of a first distance between the partial time-series within the cluster and the representative value of the measurement time-series (K-means is a clustering algorithm that divides given data points into several clusters centered on centroids … the model trainer defines centroids and determines a cluster size and a cluster radius for each cluster, fig. 2B, para. 37).
Ye et al. does not explicitly teach but Arora et al. teach wherein the measurement time-series relates to a communications network resource (detect and correct issues that can limit performance in communication systems, para. 2, collect performance data from multiple terminals over time, para. 4), and wherein the measurement time-series is collected over a predetermined timeframe (occurring at different points over a significant period of time (e.g., weeks, months, etc.), para. 5), wherein the plurality of training time-series relate to a plurality of communications network resources (with the collection of labeled data, the system trains machine learning models that can predict the classification or cluster that best represents the situation for the terminal, para. 8), wherein a cluster anomaly label is associated with said cluster (labels applied to the clusters, para. 7), wherein the cluster anomaly label encodes whether the cluster is anomalous (labels applied to the clusters, para. 7) and compute a primary anomaly label associated with the measurement time-series (each indicator reports can be classified so that the indicator reports in the repository are all labeled with the most likely network problem experienced, para. 7), wherein the primary anomaly label is computed as a function of the cluster anomaly label of the at least one cluster of the cluster subset associated with the measurement time-series (the labels applied to the cluster are then applied back to the individual examples across the full repository, so that each data set or indicator report receives the labels that is most applicable. The system also compare the set of anomalies in a report with the type of anomalies in each cluster, where the system can evaluate indicator reports to determine the types of anomalies indicated by each report, para. 7, with the collection of labeled data, the system trains machine learning models that can predict the classification or cluster that best represents the situation for the terminal, para. 8).
It would have been obvious to modify the apparatus of Ye et al. by adding Arora et al. machine learning to enhance performance. A person of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make the modification because it would aid in managing satellite terminals and other communication devices (para. 1).
In regard to claim 3, Ye et al. does not explicitly teach but Arora et al. teach the apparatus according to claim 1, wherein the apparatus is further caused to:
compute a decision weight associated to each of the at least one cluster of the cluster subset associated with the measurement time-series, wherein the decision weight depends on a similarity parameter representing similarity between a representative vector associated with the measurement time-series and the at least one cluster of the cluster subset, and on a size of the at least one cluster, wherein the size of the cluster is a number of partial time-series in the cluster (analysis are provided to a model parameter adjustment module that alters the values of parameters in the machine learning model … adjustment module can adjust the values of weights and biased, para. 94); and
compute the primary anomaly label as a function of the decision weight and the cluster anomaly label of the at least one cluster of the cluster subset associated with the measurement time-series (determine which indicator reports best fit the different clusters … determines which cluster of anomaly types is most similar to the set of anomalies present in the indicator report .. labels are then applied … can select adjustment actions as appropriate, para. 71).
Refer to claim 1 for motivational statement.
In regard to claim 4, Ye et al. does not explicitly teach but Arora et al. teach the apparatus according to claim 1, wherein the apparatus is further caused to transmit the primary anomaly label to a correction module, wherein the correction module performs root cause analysis and at least one corrective action relating to the communications network resource (system can receive and process information, detect performance limitations and other network problems and determine actions to correct or mitigate network problems, para. 46, where labels aid in this process, para. 66).
Refer to claim 1 for motivational statement.
In regard to claim 5, Ye et al. teach the apparatus according to claim 1, wherein a temporal attribute is associated with said measured time-series or each of said measurement time-series, wherein each cluster comprises a cluster temporal attribute, and wherein the external similarity condition is a function of a second distance between the cluster temporal attribute and the temporal attribute associated with the measurement time-series (K-means is a clustering algorithm that divides given data points into several clusters centered on centroids … the variance predictor to determine a metric normalized distance for the respective historical point data values, para. 37, fig. 2B).
In regard to claim 11, Ye et al. teach a method for anomaly detection, the method comprising:
collecting a measurement time-series relating to a performance indicator (the data store may store any number of tables at any point in time, the tables may be time-series point data values, para. 29);
computing a representative value of said measurement time-series, wherein the representative value is a median value of said measured time-series or each of said measurement time-series (the centroid may represent an expected or target value for the point data values such as a mean or median value of the historical point data values, para. 37);
providing a clustering model comprising a set of clusters, wherein the clustering model has been trained on a plurality of training time-series relating to the performance indicator (K-means is a clustering algorithm that divides given data points into several clustered centered on centroids, a model trainer generates one or more K-means models based on the historical point data values, para. 37), wherein a cluster of the set of clusters comprises partial time-series that meet an internal similarity condition (the historical point data values 152H may represent a selected or random subset or sampling of the point data values 152 so that the model 212 is trained using less than all of the point data values 152 that the point data anomaly detector 160 receives, para. 33), wherein the cluster of the set of clusters is defined by the values of the partial time-series and the internal similarity condition is a maximal distance value, wherein the partial time-series are portions of the plurality of training time-series, wherein the clustering model takes as input the measurement time-series (based on the determined variance value for a given input point data value, determines whether the input point data value is an anomalous point data value, para. 44-48);
selecting a cluster subset within the set of clusters, wherein the cluster subset is associated with the measurement time-series, wherein the cluster subset comprises at least one cluster which meets an external similarity condition with the measurement time-series, wherein the external similarity condition is a function of a first distance between the partial time-series within the at least one cluster and the representative value of the measurement time- series (K-means is a clustering algorithm that divides given data points into several clusters centered on centroids … the model trainer defines centroids and determines a cluster size and a cluster radius for each cluster, fig. 2B, para. 37).
Ye et al. does not explicitly teach but Arora et al. teach wherein the measurement time-series relates to a communications network resource (detect and correct issues that can limit performance in communication systems, para. 2, collect performance data from multiple terminals over time, para. 4), wherein the measurement time-series is collected over a predetermined timeframe (occurring at different points over a significant period of time (e.g., weeks, months, etc.), para. 5), wherein the plurality of training time-series relate to a plurality of communications network resources (with the collection of labeled data, the system trains machine learning models that can predict the classification or cluster that best represents the situation for the terminal, para. 8), wherein a cluster anomaly label is associated with said cluster (labels applied to the clusters, para. 7), , wherein the cluster anomaly label encodes whether the cluster is anomalous (labels applied to the clusters, para. 7), computing a primary anomaly label associated with the measurement time-series (each indicator reports can be classified so that the indicator reports in the repository are all labeled with the most likely network problem experienced, para. 7), wherein the primary anomaly label is computed as a function of the cluster anomaly label of the at least one cluster of the cluster subset associated with the measurement time-series (the labels applied to the cluster are then applied back to the individual examples across the full repository, so that each data set or indicator report receives the labels that is most applicable. The system also compare the set of anomalies in a report with the type of anomalies in each cluster, where the system can evaluate indicator reports to determine the types of anomalies indicated by each report, para. 7, with the collection of labeled data, the system trains machine learning models that can predict the classification or cluster that best represents the situation for the terminal, para. 8).
Refer to claim 1 for motivational statement.
In regard to claim 13, Ye et al. does not explicitly teach but Arora et al. teach the method according to claim 11, further comprising:
computing a decision weight associated to each of the at least one cluster of the cluster subset associated with the measurement time-series, wherein the decision weight depends on a similarity parameter representing similarity between a representative vector associated with the measurement time-series and the at least one cluster of the cluster subset, and on a size of the at least one cluster, wherein the size of the cluster is a number of partial time-series in the cluster (analysis are provided to a model parameter adjustment module that alters the values of parameters in the machine learning model … adjustment module can adjust the values of weights and biased, para. 94); and
computing the primary anomaly label as a function of the decision weight and the cluster anomaly label of the at least one cluster of the cluster subset associated with the measurement time-series (determine which indicator reports best fit the different clusters … determines which cluster of anomaly types is most similar to the set of anomalies present in the indicator report .. labels are then applied … can select adjustment actions as appropriate, para. 71).
Refer to claim 1 for motivational statement.
In regard to claim 15, Ye et al does not explicitly teach but Arora et al. teach the apparatus according to claim 1, wherein the communications network resource is associated with resource metadata, the resource metadata including attributes relating to a physical feature of the communications network resource and an environment of the communications network resource (with the clusters of related types of anomalies, the system can generate visualizations of the clusters and the relationships among them, para. 6).
Refer to claim 1 for motivational statement.
In regard to claim 16, Ye et al does not explicitly teach but Arora et al. teach the apparatus according to claim 1, wherein the performance indicator includes a network capacity, a network usage rate, a data rate, a throughput rate, a Call Setup Success rate, or a Drop Call rate (network problems they represent, such as “uplink transport problems”, “heavy download usage”, “heavy download usage with sub-optimal transport performance”, ect., para. 6).
Refer to claim 1 for motivational statement.
In regard to claim 18, Ye et al does not explicitly teach but Arora et al. teach the method according to claim 11, wherein the communications network resource is associated with resource metadata, the resource metadata including attributes relating to a physical feature of the communications network resource and an environment of the communications network resource (with the clusters of related types of anomalies, the system can generate visualizations of the clusters and the relationships among them, para. 6).
Refer to claim 1 for motivational statement.
In regard to claim 19, Ye et al does not explicitly teach but Arora et al. teach the method according to claim 11, wherein the performance indicator includes a network capacity, a network usage rate, a data rate, a throughput rate, a Call Setup Success rate, or a Drop Call rate (network problems they represent, such as “uplink transport problems”, “heavy download usage”, “heavy download usage with sub-optimal transport performance”, ect., para. 6).
Refer to claim 1 for motivational statement.
****************
Claims 2, 6-10, 12, 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ye et al. (US 2022/0382622) in further view of Arora et al. (US 2024/0073113) in further view of Suthar et al. (US 2022/0343168).
In regard to claim 2, Ye et al. teach the apparatus according to claim 1, wherein the apparatus is further caused to:
collect a plurality of measurement time-series relating to a plurality of performance indicators, wherein the plurality of measurement time-series relates to the communications network resource, wherein the plurality of measurement time-series is collected over the predetermined timeframe (the data store may store any number of tables at any point in time, the tables may be time-series point data values, para. 29), compute a respective representative value associated with each of said measurement time-series (the centroid may represent an expected or target value for the point data values such as a mean or median value of the historical point data values, para. 37);
select within the set of clusters a cluster subset associated with each of said measurement time-series, wherein the cluster subset comprises at least one cluster for which the partial time-series within the cluster meet a distance condition with the representative value associated with the measurement time-series (K-means is a clustering algorithm that divides given data points into several clusters centered on centroids … the model trainer defines centroids and determines a cluster size and a cluster radius for each cluster, fig. 2B, para. 37).
Ye et al. does not explicitly teach but Arora et al. teach compute a primary anomaly label associated with each of said measurement time-series, wherein the primary anomaly label is computed as a function of the cluster anomaly label of the at least one cluster of the cluster subset associated with the measurement time-series (the labels applied to the cluster are then applied back to the individual examples across the full repository, so that each data set or indicator report receives the labels that is most applicable. The system also compare the set of anomalies in a report with the type of anomalies in each cluster, where the system can evaluate indicator reports to determine the types of anomalies indicated by each report, para. 7, with the collection of labeled data, the system trains machine learning models that can predict the classification or cluster that best represents the situation for the terminal, para. 8).
It would have been obvious to modify the apparatus of Ye et al. by adding Arora et al. machine learning to enhance performance. A person of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make the modification because it would aid in managing satellite terminals and other communication devices (para. 1).
Ye et al. and Arora et al. does not explicitly teach but Suthar et al. teach compute a secondary anomaly label associated with at least one of the plurality of measurement time-series, wherein the secondary anomaly label is computed as a function of the primary anomaly labels associated with the plurality of measurement time-series (generates a predicted value for the second data point based on processing data points in the first cluster using a first machine learning model .. labels the second data points as anomalous, para. 72, fig. 7).
It would have been obvious to modify the apparatus of Ye et al. and Arora et al. by adding Suthar et al. real-time adaptive threshold. A person of ordinary skill in the art before the effective filing date of the claimed invention would have been motivated to make the modification because it would aid in real-time monitoring and anomaly detection (para. 72).
In regard to claim 6, Ye et al. teach the apparatus according to claim 1, wherein the apparatus is further caused to:
collect a plurality of measurement time-series relating to the plurality of communications network resources associated with the plurality of communications network resources (the data store may store any number of tables at any point in time, the tables may be time-series point data values, para. 29), wherein the feature vector,
encode physical features of the plurality of communications network resources (K-means model provides cluster based anomaly detection approach and supports geography features, para. 37);
compute representative values associated with each measurement time-series of the time-series subset (the centroid may represent an expected or target value for the point data values such as a mean or median value of the historical point data values, para. 37);
select within the set of clusters a cluster subset associated with each measurement time-series of the time-series subset, wherein the cluster subset comprises at least one cluster which meets the external similarity condition with the measurement time-series (K-means is a clustering algorithm that divides given data points into several clusters centered on centroids … the model trainer defines centroids and determines a cluster size and a cluster radius for each cluster, fig. 2B, para. 37).
Ye et al. does not explicitly teach but Arora et al. teach collect feature vectors associated with the plurality of communication network resources (for each reported set of performance information from a terminal, the system can generate a vector that includes the classifications determined for the respective metrics reported by the terminal, para. 4), select a time-series subset within the plurality of measurement time-series as a function of the feature vectors, wherein the feature vectors associated to the measurement time-series within the time-series subset meet a similarity criterion (the record includes a high-dimensional vector of indicators corresponding to particular terminal in a particular time interval … group together to create a peer group where all terminals shares the same outroute or service plan during a given time interval, para. 74); compute a primary anomaly label associated with each measurement time-series of the time-series subset (each indicator reports can be classified so that the indicator reports in the repository are all labeled with the most likely network problem experienced, para. 7), wherein the primary anomaly label is computed as a function of the cluster anomaly label of the at least one cluster of the cluster subset associated with the measurement time-series (the labels applied to the cluster are then applied back to the individual examples across the full repository, so that each data set or indicator report receives the labels that is most applicable. The system also compare the set of anomalies in a report with the type of anomalies in each cluster, where the system can evaluate indicator reports to determine the types of anomalies indicated by each report, para. 7, with the collection of labeled data, the system trains machine learning models that can predict the classification or cluster that best represents the situation for the terminal, para. 8).
Refer to claim 2 for motivational statement.
Ye et al. and Arora et al. does not explicitly teach but Suthar et al. teach compute a secondary anomaly label associated with at least one of the plurality of measurement time-series, wherein the secondary anomaly label is computed as a function of the primary anomaly labels associated with the plurality of measurement time-series (generates a predicted value for the second data point based on processing data points in the first cluster using a first machine learning model .. labels the second data points as anomalous, para. 72, fig. 7).
Refer to claim 2 for motivational statement.
In regard to claim 7, Ye et al. does not explicitly teach but Martyanov teaches the apparatus according to claim 6, wherein said similarity criterion consists in that the feature vectors associated to the measurement time-series within the time-series subset are identical (the record includes a high-dimensional vector of indicators corresponding to particular terminal in a particular time interval … group together to create a peer group where all terminals shares the same outroute or service plan during a given time interval, para. 74)Refer to claim 2 for motivational statement.
Refer to claim 2 for motivational statement.
In regard to claim 8, Ye et al. teach the apparatus according to claim 6, wherein providing the clustering model comprises: providing a training array comprising the plurality of training time-series within the time-series subset as columns of the training array and simultaneous values of the training time-series within the time-series subset as lines of the training array, wherein a timestamp is associated to each line of the training array (any number of point data values may be time-series point data values, para. 29, a model trains on point data values, para. 31).
Ye et al. does not explicitly teach but Arora et al. teach clustering the lines of the training array into at least one subarray, wherein the at least one subarray comprises lines of the training array which meet a first vector similarity condition, wherein the at least one subarray comprises a partial column corresponding to each column of the training array (the training data includes many examples where each example represents a feature vector and a label based on a cluster that is most similar to the anomalies in the feature vector, para. 87); and associating the cluster anomaly labels with the set of clusters (each example in the training data can have a corresponding label indicating a classification, selected from among various predetermined classes or categories that is believed to best describe the state of the terminal, para. 89).
Refer to claim 2 for motivational statement.
Ye et al. and Arora et al. does not explicitly teach but Suthar et al. teach clustering the partial columns of the at least one subarray into said set of clusters, wherein each cluster of the set of clusters comprises partial columns of the at least one subarray that meet a second vector similarity condition (generates a predicted value for the second data point based on processing data points in the first cluster using a first machine learning model .. labels the second data points as anomalous, para. 72, fig. 7).
Refer to claim 2 for motivational statement.
In regard to claim 9, Ye et al. and Arora et al. does not explicitly teach but Suthar et al. teach the apparatus according to claim 1, wherein the apparatus is further caused to set the cluster anomaly label associated with the cluster to encode an anomalous cluster in response to determining that a size of the cluster is lower than a threshold of size (adaptive threshold … the Cleansing Component receives data points sequentially and store them until a predefined number of points have been collected to proceed as a batch, para. 42, or timestamps are within a predefined threshold, para. 44).
Refer to claim 2 for motivational statement.
In regard to claim 10, Ye et al. and Arora et al. does not explicitly teach but Suthar et al. teach the apparatus according to claim 9, wherein the apparatus is further caused to set the threshold of size as a function of a size distribution of the set of clusters (control the number of data points to utilize, para. 45).
Refer to claim 2 for motivational statement.
In regard to claim 12, Ye et al. teach the method according to claim 11, the method comprising:
collecting a plurality of measurement time-series relating to a plurality of performance indicators, wherein the plurality of measurement time-series relates to the communications network resource, wherein the plurality of measurement time-series is collected over the predetermined timeframe (the data store may store any number of tables at any point in time, the tables may be time-series point data values, para. 29), compute a respective representative value associated with each of said measurement time-series (the centroid may represent an expected or target value for the point data values such as a mean or median value of the historical point data values, para. 37);
computing a respective representative value associated with each of said measurement time-series (the centroid may represent an expected or target value for the point data values such as a mean or median value of the historical point data values, para. 37);
selecting within the set of clusters a cluster subset associated with each of said measurement time-series, wherein the cluster subset comprises at least one cluster for which the partial time-series within the cluster meet a distance condition with the representative value associated with the measurement time-series (K-means is a clustering algorithm that divides given data points into several clusters centered on centroids … the model trainer defines centroids and determines a cluster size and a cluster radius for each cluster, fig. 2B, para. 37).
Ye et al. does not explicitly teach but Arora et al. teach computing a primary anomaly label associated with each of said measurement time-series, wherein the primary anomaly label is computed as a function of the cluster anomaly label of the at least one cluster of the cluster subset associated with the measurement time-series (the labels applied to the cluster are then applied back to the individual examples across the full repository, so that each data set or indicator report receives the labels that is most applicable. The system also compare the set of anomalies in a report with the type of anomalies in each cluster, where the system can evaluate indicator reports to determine the types of anomalies indicated by each report, para. 7, with the collection of labeled data, the system trains machine learning models that can predict the classification or cluster that best represents the situation for the terminal, para. 8).
Refer to claim 2 for motivational statement.
Ye et al. and Arora et al. does not explicitly teach but Suthar et al. teach computing a secondary anomaly label associated with at least one of the plurality of measurement time-series, wherein the secondary anomaly label is computed as a function of the primary anomaly labels associated with the plurality of measurement time-series (generates a predicted value for the second data point based on processing data points in the first cluster using a first machine learning model .. labels the second data points as anomalous, para. 72, fig. 7).
Refer to claim 2 for motivational statement.
In regard to claim 14, Ye et al. teach the method according to claim 11, further comprising:
collecting a plurality of measurement time-series relating to the plurality of communications network resources associated with the plurality of communications network resources (the data store may store any number of tables at any point in time, the tables may be time-series point data values, para. 29),
compute representative values associated with each measurement time-series of the time-series subset (the centroid may represent an expected or target value for the point data values such as a mean or median value of the historical point data values, para. 37);
selecting within the set of clusters a cluster subset associated with each measurement time-series of the time-series subset, wherein the cluster subset comprises at least one cluster which meets the external similarity condition with the measurement time-series (K-means is a clustering algorithm that divides given data points into several clusters centered on centroids … the model trainer defines centroids and determines a cluster size and a cluster radius for each cluster, fig. 2B, para. 37).
Ye et al. does not explicitly teach but Arora et al. teach collecting feature vectors associated with the plurality of communication network resources (for each reported set of performance information from a terminal, the system can generate a vector that includes the classifications determined for the respective metrics reported by the terminal, para. 4), wherein the feature vectors encode physical features of the plurality of communications network resources (feature vector indicates various performance indicators, para. 88); selecting a time-series subset within the plurality of measurement time-series as a function of the feature vectors, wherein the feature vectors associated to the measurement time-series within the time-series subset meet a similarity criterion (the record includes a high-dimensional vector of indicators corresponding to particular terminal in a particular time interval … group together to create a peer group where all terminals shares the same outroute or service plan during a given time interval, para. 74); and computing a primary anomaly label associated with each measurement time-series of the time-series subset, wherein the primary anomaly label is computed as a function of the cluster anomaly label of the at least one cluster of the cluster subset associated with the measurement time-series (the labels applied to the cluster are then applied back to the individual examples across the full repository, so that each data set or indicator report receives the labels that is most applicable. The system also compare the set of anomalies in a report with the type of anomalies in each cluster, where the system can evaluate indicator reports to determine the types of anomalies indicated by each report, para. 7, with the collection of labeled data, the system trains machine learning models that can predict the classification or cluster that best represents the situation for the terminal, para. 8).
Refer to claim 2 for motivational statement.
Ye et al. and Arora et al. does not explicitly teach but Suthar et al. teach computing a secondary anomaly label associated with at least one measurement time-series of the time-series subset, wherein the secondary anomaly label is computed as a function of the primary anomaly labels associated with the measurement time-series of the time-series subset (generates a predicted value for the second data point based on processing data points in the first cluster using a first machine learning model .. labels the second data points as anomalous, para. 72, fig. 7).
Refer to claim 2 for motivational statement.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. See PTO 892.
Xue et al. (US 2020/0167639) model training
ABDI TAGHI ABAD et al. (US 2020/0285889) learning of neural networks
Wang et al. (US 2020/0334539) multi-model structures for classification
Trinh et al. (US 2020/0379454) machine learning based predictive maintenance
*************
Schmitt et al. (US 12,093,818) time-series data
Sharpe et al. (US 2024/0119303) labels, anomalous clusters
Arora et al. (US 2024/0073113) cluster anomaly types and applied labels
Saeed et al. (US 2023/0125203) anomaly detection
Wong et al. (US 2023/0118240) weights, training, label for anomaly
Lu et al. (US 2022/0004843) anomaly detection with weight, vector, label
Chivu et al. (US 12,238,119) anomalous events models
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 LOAN TRUONG whose telephone number is 408-918-7552. The examiner can normally be reached on 10AM-6PM PST M-F.
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, Thomas Ashish can be reached on 571-272-0631. 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.
/Loan L.T. Truong/Primary Examiner, Art Unit 2114 Loan.truong@uspto.gov
Prosecution Insights