Prosecution Insights
Last updated: November 04, 2026
Application No. 18/387,571

METHOD FOR PROCESSING MISSING VALUE IN NETWORK LOG DATA AND CLASSIFYING ROOT CAUSE OF COMMUNICATION DEFECT THEREBY

Non-Final OA §103§112
Filed
Nov 07, 2023
Priority
Nov 10, 2022 — RE 10-2022-0149179
Examiner
BAKER, EZRA JAMES
Art Unit
2126
Tech Center
2100 — Computer Architecture & Software
Assignee
Innowireless Co. Ltd.
OA Round
2 (Non-Final)
47%
Grant Probability
Moderate
2-3
OA Rounds
1y 2m
Est. Remaining
66%
With Interview

Examiner Intelligence

Grants 47% of resolved cases
47%
Career Allowance Rate
14 granted / 30 resolved
-8.3% vs TC avg
Strong +20% interview lift
Without
With
+19.6%
Interview Lift
resolved cases with interview
Typical timeline
4y 2m
Avg Prosecution
16 currently pending
Career history
55
Total Applications
across all art units

Statute-Specific Performance

§101
32.1%
-7.9% vs TC avg
§103
37.7%
-2.3% vs TC avg
§102
8.3%
-31.7% vs TC avg
§112
20.4%
-19.6% vs TC avg
Black line = Tech Center average estimate • Based on career data from 30 resolved cases

Office Action

§103 §112
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 present application is being examined under the claims filed 08/24/2026. Claims 1-12 are pending. Response to Amendment This Office Action is in response to Applicant’s communication filed 08/24/2026 in response to office action mailed 05/22/2026. The Applicant’s remarks and any amendments to the claims or specification have been considered with the results that follow. Response to Arguments In Remarks pages 5-7, Argument 1 (Examiner summarizes Applicant’s arguments) Applicant argues that the objections and rejections under 35 U.S.C. 112(b) have been overcome by the amendments. Examiner’s response to Argument 1, The objections and most of the rejections under 35 U.S.C. 112(b) have been overcome by Applicant’s amendments. However, some rejections under 35 U.S.C. 112(b) remain and the amendments require new rejections under 35 U.S.C. 112(a) for including new matter. In Remarks pages 7-12, Argument 2 (Examiner summarizes Argument 2) Applicant traverses the rejections under 35 U.S.C. 103 arguing that Kim and Cini do not teach the corresponding claim limitations, and that the references used for the dependent claims fail to cure the deficiencies of the independent claim. Examiner’s response to Argument 2 The grounds of rejection have been updated rendering applicant’s arguments moot. In Remarks page 10, Argument 3 Zhang, however, merely teaches integrating absent feature imputing and GMM clustering into a learning procedure. Zhang does not teach or suggest using the imputing and GMM clustering learning procedure to impute missing values for non-time series data among parameters other than the parameters processed in (b) imputing, i.e., the certain kinds of parameters in the non-time series data and time series data. Examiner’s response to Argument 3 Examiner points to algorithm 1 of Zhang which uses parameters for gaussian mixture models that were not used in the other imputing PNG media_image1.png 684 1054 media_image1.png Greyscale Claim Rejections - 35 USC § 112(a) The following is a quotation of the first paragraph of 35 U.S.C. 112(a): (a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention. The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112: The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention. Claims 3, 5, 6, and 11 are rejected under 35 U.S.C. 112(a) for failing to provide a written description of the claimed invention. Regarding Claim 3 Claim 3 recites “for Radio Frequency-related (RF-related) parameters”. However, the specification only discloses (page 5) “For RF-related parameters, including Reference Signals Received Power (RSRP), Reference Signal Received Quality (RSRQ), and Signal to Interference plus Noise Ratio (SINR), exclude any data samples with missing values. If a missing value in the Packet Data Convergence Protocol/Radio Link Control statistics summary parameter has the same meaning as zero, the missing value is imputed with zero.” However, the specification does not disclose radio frequency-related parameters. The specification does not convey (MPEP 2163 I B.) “with reasonable clarity to those skilled in the art that, as of the filing date sought, the inventor was in possession of the invention as now claimed. See, e.g., Vas-Cath, Inc., 935 F.2d at 1563-64, 19 USPQ2d at 1117.” Regarding Claim 5 Claim 5 recites characterized in that the parameters other than those processed in (b) imputing include Primary Cell (PCell) Estimated Distance, Key Performance Indicator Primary Cell Physical Downlink Shared Channel (KPI PCell PDSCH) & Physical Uplink Shared Channel Block Error Rate Percentage (PUSCH BLER[%]), Physical Layer Residual Block Error Rate (PHY R-BLER} Info Physical Downlink Shared Channel (PDSCH) Total Info Down Link Residual Block Error Rate Percentage (DL R-BLER[%]), and Total Info Physical Layer Residual Block Error Rate (PHY R-BLER) Info Physical Uplink Shared Channel Residual Block Error Rate Percentage (PUSCH UL R-BLER[%]). The specification does not disclose any of the newly amended (underlined) portions, but instead discloses “Parameters other than those processed in operation (b) include PCell Estimated Distance, KPI PCell PDSCH & PUSCH BLER[%], PHY R-BLER Info PDSCH Total Info DL R-BLER[%], and Total Info PHY R-BLER Info PUSCH UL R-BLER[%].” The specification does not convey (MPEP 2163 I B.) “with reasonable clarity to those skilled in the art that, as of the filing date sought, the inventor was in possession of the invention as now claimed. See, e.g., Vas-Cath, Inc., 935 F.2d at 1563-64, 19 USPQ2d at 1117.” Regarding Claim 6 Claim 6 recites (c3) clustering the missing value dataset (Dn) into a plurality of second clusters based on the clustering result of (c2) clustering and finding one of the second clusters for each of the values of the non-missing value dataset (Dy); This limitation is not disclosed in the specification. The specification does disclose (page 5-6) Operation (c) may include operation (c1) of dividing the non-series data into a dataset without missing value (Dy; non-missing value dataset) and a dataset with missing value (Dn; missing value dataset); operation (c2) of clustering the non-missing value dataset (Dy) through a GMM-based EM algorithm, finding the centroid of each cluster, and assigning each instance to the cluster with the closest Euclidean distance; operation (c3) of finding a cluster for each instance after clustering the missing value dataset (Dn) based on the clustering result of operation (c2) The specification does not disclose finding clusters derived from the Dn dataset for the values of the Dy dataset. Therefore, the specification does not convey (MPEP 2163 I B.) “with reasonable clarity to those skilled in the art that, as of the filing date sought, the inventor was in possession of the invention as now claimed. See, e.g., Vas-Cath, Inc., 935 F.2d at 1563-64, 19 USPQ2d at 1117.” Regarding dependent claims Claim 11 is dependent upon claim 6, and is therefore similarly rejected for including the deficiencies of claim 6. Claim Rejections - 35 USC § 112(b) The following is a quotation of 35 U.S.C. 112(b): (b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention. The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph: The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention. Claims 4 and 8-10 are rejected under 35 U.S.C. 112(b) as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention. Regarding Claim 4 Claim 4 recites the limitation "if a missing value in the Packet Data Convergence Protocol/Radio Link Control statistics summary parameter has the same meaning as zero, the missing value is imputed with zero" in line 1 of the claim. There is insufficient antecedent basis for this limitation in the claim. In particular, there is insufficient antecedent basis for the term “the same meaning”. For purposes of examination, the examiner interprets the limitation as though it said "if a missing value in the Packet Data Convergence Protocol/Radio Link Control statistics summary parameter has a Regarding Claim 8 Claim 8 recites The method of claim 7, characterized in that, (d) imputing comprises: (d1) artificially masking a predetermined percentage of data at random from the existing time series data with the missing values to generate virtual missing values; (d2) calculating alternative loss values between missing values artificially generated by mean absolute error (MAE) measurements and the imputation values after the imputation model imputes all the missing values; and (d3) reconstructing observed data present in the existing data through modeling processing and calculating the difference between the observed data and the reconstructed data by the MAE measurements to obtain loss values Regarding dependent claims Claims 9-10 are dependent upon claim 8, and are therefore similarly rejected for including the deficiencies of claim 8. Claim Rejections - 35 USC § 103 The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action: A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made. Claims 1 and 6 are rejected under 35 U.S.C. 103 as being unpatentable over NPL reference Tavazzi et al. “Exploiting mutual information for the imputation of static and dynamic mixed-type clinical data with an adaptive k-nearest neighbours approach” herein referred to as Tavazzi, and Nikolova-Simons et al. (PGPUB no. US20200372982A1). Regarding Claim 1 Tavazzi teaches: (a) separating […] into non-time series and time series data; (page 6 column 1 last paragraph) “Figure 1(a) illustrates the sample construction for subject i, with p being the number of static features, m the number of the dynamic ones, and n the number of his/her visits in the first three months of screening.” (b) imputing the missing values according to certain kinds of parameters in the non-time series data and time series data; (page 7 column 2 second to last paragraph) “At this point, a filtering step is performed: candidates with a number of comparable features with subject i smaller than the 90% of the total number of non-missing features in sample i (both computed with the same adjustment for the static features) are dropped. Once the distances to all the candidates have been computed, the k nearest ones are selected and their values in correspondence to the feature to be imputed are used for the imputation: for continuous and ordinal features, after removing possible outliers (values outside 1.5 times the interquartile range above the upper quartile and below the lower quartile), the missing feature in i is imputed with the average of the selected values, each weighted by the inverse of the corresponding candidate distance; for categorical features, the missing feature in i is imputed with the mode of the selected values.” Tavazzi does not explicitly teach: A method for processing missing values in network log data, the method comprising: network communication log data collected in a call interval (c) imputing the missing values via a Gaussian mixture model (GMM) for the non-time series data among parameters other than the parameters processed in (b) imputing; and (d) imputing the missing values via a joint-approach for the time series data among the parameters other than the parameters processed in (b) imputing. However, Zhang teaches: (c) imputing the missing values via a Gaussian mixture model (GMM) for the non-time series data among the parameters other than the parameters processed in (b) imputing; (page 1 abstract) “Specifically, the missing data is filled by the result of GMM clustering, and the imputed data is then taken for GMM clustering. These two steps alternatively negotiate with each other to achieve optimum” Tavazzi, Zhang, and the instant application are analogous because they are all directed to data processing. It would have been obvious to a person having ordinary skill in the art before the effective filing date of the present invention to modify the data separation and imputation technique of Tavazzi by additionally using Gaussian mixture model imputation taught by Zhang because (Zhang page 2 last paragraph bullet 1) “Missing features are dynamically imputed to better serve the clustering tasks, while existing observed entries are guaranteed to remain unchanged.” And Nikolova-Simons teaches: A method for processing missing values in network log data, the method comprising: (paragraph 8) “The instructions cause the processor to: evaluate records stored on a storage device of a call center of a system, wherein the storage device is configured to store electronic records of calls from a user site to the call center[*Examiner notes: network log data], and each record includes at least a situation attribute field and an outcome attribute field, identify records of the stored records which include, at least one of, a value in the situation attribute or case note text in a case note database, but no value in the outcome attribute[*Examiner notes: missing values]” network communication log data collected in a call interval (paragraph 8) “The instructions cause the processor to: evaluate records stored on a storage device of a call center of a system, wherein the storage device is configured to store electronic records of calls from a user site to the call center[*Examiner notes: network communication log data in a call interval], and each record includes at least a situation attribute field and an outcome attribute field” and (d) imputing the missing values via a joint-approach for the time series data among the parameters other than the parameters processed in (b) imputing. [*Examiner notes: The broadest reasonable interpretation of the term “parameters” includes any data being used in a calculation or other operation.]; (paragraph 38) “The situation attribute 400 either identifies a reason for the call via a structured situation string or specifies that there is a case note with unstructured narrative text for the call[*Examiner notes: parameters other than parameters processed in (b) imputing]. FIG. 4 shows example situation attributes 400 for the type incident and includes bleeding/injury 402, anxious 406, breathing problem 408, dizziness 410, request responder 412, and see case note 414”; (paragraph 40) “As described in greater detail below, the missing outcome instruction set or algorithm 218 includes instructions, which, when executed by the processor 206, cause the processor 206 to at least identify records with a situation attribute or an associated case note but no outcome attribute, and imputes an outcome attribute to the outcome field of the identified record using either the situation attribute or the case note if available.”; (paragraph 49) “If there is more than one predicted value, then at 612, the predicted values are combined or blended according to predetermined assessment rules to produce a single imputed outcome[*Examiner notes: joint approach] for the file. ” Tavazzi, Zhang, Nikolova-Simons, and the instant application are analogous because they are all directed to data processing. It would have been obvious to a person having ordinary skill in the art before the effective filing date of the present invention to modify the data separation and imputation as taught by Tavazzi in view of Zhang by applying the techniques to the network call log data of Nikolova-Simons because (Nikola-Simons paragraph 6) “The predictive models are trained and tested with the outcome attribute of records. However, the attribute fields in each record for each call from the user site 104 to the call center 106 are not always populated” and it would have been further obvious to additionally use the joint imputation approach as taught by Nikolova-Simons because (Nikolova-Simons paragraph 41) “By also using imputed outcomes, the predictive analytics instruction set or algorithm 220 uses additional information that was not previously available, which can reduce patient risk, lower costs and/or improve patient outcomes.” Regarding Claim 6 Tavazzi in view of Zhang and Nikolova-Simons teaches: The method of claim 1 (see rejection of claim 1) Zhang further teaches: characterized in that, (c) imputing comprises: (c1) dividing the non-series data into a dataset without missing value (Dy; non-missing value dataset) and a dataset with missing value (Dn; missing value dataset);(page 6 last paragraph) “When we optimize the prior of k components, the mean μk, and the covariance Σk for the kth Gaussian basis, we propose to optimize the missing parts of the data xj(mj)[*Examiner notes: ] while keeping the observable features xj(oj)[*Examiner notes: non-missing value] unchanged during the optimization process” (c2) clustering the non-missing value dataset (Dy) through a GMM-based EM algorithm into a plurality of first clusters, finding a centroid of each of the first clusters, and assigning each of the values of the non-missing value dataset (Dy) to the cluster with a closest Euclidean distance (page 6 section 3.2.3) PNG media_image2.png 116 566 media_image2.png Greyscale (c3) clustering the missing value dataset (Dn) into a plurality of second clusters based on the clustering result of (c2) clustering and finding one of the second clusters for each of the values of the non-missing value dataset (Dy); (c4) finding a complete dataset that is closest to the dataset with missing values based on the Euclidean distance from each of the second clusters found in (c3) clustering; and (c5) imputing the missing values with an average value of the complete dataset found in (c4) finding. (page 7 below equation 14) “As seen from Equation (14), the missing elements of each sample xj are imputed with the corresponding dimension of the Gaussian mixture component and the observable features of the sample.”; [*Examiner notes: Algorithm 1 and section 3.2 describes the clustering method in detail, which is consistent with what is claimed.] It would have been obvious to a person having ordinary skill in the art before the effective filing date of the present invention to combine Kim and Cini with Zhang for the same reasons given in claim 1 above. Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Tavazzi in view of Zhang, Nikolova-Simons, and further in view of Koziol et al (PGPUB no US 20210211899 A1) herein referred to as Koziol. Regarding Claim 2 Tavazzi in view of Zhang and Nikolova-Simons teaches: The method of claim 1 (see rejection of claim 1) Tavazzi in view of Zhang and Nikolova-Simons does not teach: characterized in that the Serving Physical Cell ID (PCI) parameter creates a new parameter to represent it, with a value of 1 if data is present on the Serving PCI and 0 if data is missing. However, Koziol teaches: characterized in that the Serving Physical Cell ID (PCI) parameter creates a new parameter to represent it, with a value of 1 if data is present on the Serving PCI and 0 if data is missing. (paragraph [0002]) “Reported CGIs may be used to create neighbor relationships for cells associated with a physical cell identifier (PCI), which may be reported by user equipment to a network entity that is missing an entry in a neighbor relation table (NRT), as illustrated in FIG. 1.” Tavazzi, Zhang, Nikolova-Simons, Koziol, and the instant application are analogous because they are all directed to data processing. It would have been obvious to a person having ordinary skill in the art before the effective filing date of the present invention to modify the network communication log and imputations of Tavazzi in view of Zhang and Nikolova-Simons with the physical cell ID parameters as taught by Koziol because (Koziol paragraph [0002]) “Automatic neighbor relation (ANR) functions relieve a wireless network operator from the burdens of manually managing neighbor cell relation (NCR) lists. NCR lists provide a mapping between a cell global identity (CGI) and a physical cell identity (PCI) and its associated frequencies. CGIs are also created for each cell within a network entity, such as an evolved Node B (eNB) or next generation Node B (gNB).” Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Tavazzi in view of Zhang and Nikolova-Simons, and further in view of Cha et al. (PGPUB no. US 20210376894 A1) herein referred to as Cha. Regarding Claim 3 Tavazzi in view of Zhang and Nikolova-Simons teaches: The method of claim 1 (see rejection of claim 1) Tavazzi in view of Zhang and Nikolova-Simons does not teach: characterized in that, for RF-related parameters including Reference Signals Received Power (RSRP), Reference Signal Received Quality (RSRQ), and Signal to Interference plus Noise Ratio (SINR), data samples with missing values is excluded if they exist. However, Cha teaches: characterized in that, for Radio Frequency-related (RF-related) parameters including Reference Signals Received Power (RSRP), Reference Signal Received Quality (RSRQ), and Signal to Interference plus Noise Ratio (SINR), data samples with missing values is excluded if they exist. (paragraph [0485]) “For example, the UE may ignore/exclude the measurement value (e.g., RSRP, RSRQ, SINR, etc.) acquired before the new beam setting reporting/request time. In other words, ignoring/excluding the measurement values (e.g., RSRP, RSRQ, and SINR) may be interpreted as excluding the measurement values in the process of calculating/determining an average value of measurement values acquired over multiple periodicities in order to report the RS index(es) and/or RS index(es) and measurement values (e.g., RSRP, RSRQ, and SINR).” Tavazzi, Zhang, Nikolova-Simons, Cha, and the instant application are analogous because they are all directed to data processing. It would have been obvious to a person having ordinary skill in the art before the effective filing date of the present invention to modify the network communication log and imputations of Tavazzi in view of Zhang and Nikolova-Simons with the exclusion of certain parameters as taught by Cha because (Cha paragraph [0485]) “ignoring/excluding the measurement values (e.g., RSRP, RSRQ, and SINR) may be interpreted as excluding the measurement values in the process of calculating/determining an average value of measurement values acquired over multiple periodicities in order to report the RS index(es) and/or RS index(es) and measurement values (e.g., RSRP, RSRQ, and SINR).” Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Tavazzi in view of Zhang, Nikolova-Simons, and further in view of NPL reference Kim et al. “Unsupervised Anomaly Detection and Root Cause Analysis in Mobile Networks” herein referred to as Kim. Regarding Claim 4 Tavazzi in view of Zhang and Nikolova-Simons teaches: The method of claim 1 (see rejection of claim 1) Tavazzi in view of Zhang and Nikolova-Simons does not explicitly teach: characterized in that, if a missing value in the Packet Data Convergence Protocol/Radio Link Control statistics summary parameter has the same meaning as zero, the missing value is imputed with zero. However, Kim teaches: characterized in that, if a missing value in the Packet Data Convergence Protocol/Radio Link Control statistics summary parameter has the same meaning as zero, the missing value is imputed with zero. (page 178 column 2 last paragraph) “By taking the frequency or median values of the features, the presence of null values can be neglected unless all the records being aggregated have null values. In such cases, we impute null values with zero for categorical features[*Examiner notes: certain kinds of parameters].” Tavazzi, Zhang, Nikolova-Simons, Kim, and the instant application are analogous because they are all directed to data processing. It would have been obvious to a person having ordinary skill in the art before the effective filing date of the present invention to modify the network communication log and imputations of Tavazzi in view of Zhang and Nikolova-Simons with the zero imputation of Kim because (Kim page 178 column 2 last paragraph) “By taking the frequency or median values of the features, the presence of null values can be neglected unless all the records being aggregated have null values. In such cases, we impute null values with zero for categorical features. Since the aggregated values of categorical features count the frequency of each category, this imputation is justified.” Claim 5 is rejected under 35 U.S.C. 103 as being unpatentable over Tavazzi in view of Zhang and Nikolova-Simons, and further in view of NPL reference Chen et al. “Cellular network location estimation via RSS-based data clean enhanced scheme” herein referred to as RSS. Regarding Claim 5 Tavazzi in view of Zhang and Nikolova-Simons teaches: The method of claim 1 (see rejection of claim 1) Tavazzi in view of Zhang and Nikolova-Simons does not explicitly teach: characterized in that the parameters other than those processed in (b) imputing include Primary Cell (PCell) Estimated Distance, Key Performance Indicator Primary Cell Physical Downlink Shared Channel (KPI PCell PDSCH) & Physical Uplink Shared Channel Block Error Rate Percentage (PUSCH BLER[%]}, Physical Layer Residual Block Error Rate (PHY R-BLER} Info Physical Downlink Shared Channel (PDSCH)_Total Info Down Link Residual Block Error Rate Percentage (DL R-BLER[%]j, and Total Info Physical Layer Residual Block Error Rate (PHY R-BLER} Info Physical Uplink Shared Channel Residual Block Error Rate Percentage (PUSCH UL R-BLER[%]). However, RSS teaches: characterized in that the parameters other than those processed in (b) imputing include Primary Cell (PCell) Estimated Distance, Key Performance Indicator Primary Cell Physical Downlink Shared Channel (KPI PCell PDSCH) & Physical Uplink Shared Channel Block Error Rate Percentage (PUSCH BLER[%]}, Physical Layer Residual Block Error Rate (PHY R-BLER} Info Physical Downlink Shared Channel (PDSCH)_Total Info Down Link Residual Block Error Rate Percentage (DL R-BLER[%]j, and Total Info Physical Layer Residual Block Error Rate (PHY R-BLER} Info Physical Uplink Shared Channel Residual Block Error Rate Percentage (PUSCH UL R-BLER[%]). (page 930 column 1 second to last paragraph) “A RSS-based data clean scheme enhanced empirical learning algorithm for cellular mobile terminals was developed and tested. We presented the details of the scheme and how it constructs the remaining set based on joints of predicted distance spheres.” Tavazzi, Zhang, Nikolova-Simons, RSS, and the instant application are analogous because they are all directed to data processing. It would have been obvious to a person having ordinary skill in the art before the effective filing date of the present invention to modify the network communication log and imputations of Tavazzi in view of Zhang and Nikolova-Simons by using the parameters of RSS because (RSS page 930 column 1 second to last paragraph) “Our results show that, compared with the conventional coordinates based SVR, DCSEEL can provide accuracy improvement with an average value of 47.17% in urban areas, 46.95% in suburban areas, and 17.74% in rural areas.” Claims 7-9 are rejected under 35 U.S.C. 103 as being unpatentable over Tavazzi in view of Zhang, Nikolova-Simons, and further in view of NPL reference Cini et al. “FILLING THE G_AP_S: MULTIVARIATE TIME SERIES IMPUTATION BY GRAPH NEURAL NETWORKS” herein referred to as Cini. Regarding Claim 7 Tavazzi in view of Zhang and Nikolova-Simons teaches: The method of claim 1 (see rejection of claim 1) Tavazzi in view of Zhang and Nikolova-Simons does not explicitly teach: characterized in that (d) imputing imputes the missing value in a way that minimizes a first loss of reconstructing the existing time series data with the missing values and a second loss of randomly imputing the missing value by iteratively training to minimize the first and second losses. However, Cini teaches: characterized in that (d) imputing imputes the missing value in a way that minimizes a first loss of reconstructing the existing time series data with the missing values and a second loss of randomly imputing the missing value by iteratively training to minimize the first and second losses. (page 4 just above equation 1) PNG media_image3.png 314 553 media_image3.png Greyscale Tavazzi, Zhang, Nikolova-Simons, Cini, and the instant application are analogous because they are all directed to data processing. It would have been obvious to a person having ordinary skill in the art before the effective filing date of the present invention to modify the network communication log and imputations of Tavazzi in view of Zhang and Nikolova-Simons with the loss optimization as taught by Cini because (Cini page 1 abstract) “In particular, we introduce a novel graph neural network architecture, named GRIN, which aims at reconstructing missing data in the different channels of a multivariate time series by learning spatio-temporal representations through message passing. Empirical results show that our model outperforms state-of-the-art methods in the imputation task on relevant real-world benchmarks with mean absolute error improvements often higher than 20%” Regarding Claim 8 Tavazzi in view of Zhang, Nikolova-Simons, and Cini teaches: The method of claim 7 (see rejection of claim 7) Cini further teaches: characterized in that, (d) imputing comprises: (d1) artificially masking a predetermined percentage of data at random from the existing time series data with the missing values to generate virtual missing values; (d2) calculating alternative loss values between missing values artificially generated by the mean absolute error (MAE) measurements and the imputation values after the imputation model imputes all the missing values; and (d3) reconstructing observed data present in the existing data through modeling processing and calculating the difference between the observed data and the reconstructed data by the MAE measurements to obtain loss values. (page 4 just above equation 1) PNG media_image3.png 314 553 media_image3.png Greyscale It would have been obvious to a person having ordinary skill in the art before the effective filing date of the present invention to combine Tavazzi, Zhang, and Nikolova-Simons with Cini for the same reasons given in claim 7 above. Regarding Claim 9 Tavazzi in view of Zhang, Nikolova-Simons, and Cini teaches: The method of claim 8 (see rejection of claim 8) And Cini further teaches: characterized in that, in order to recognize the virtual missing values, an Indicating Mask that denotes artificially masked values as 1 and all other values as 0 is utilized. (page 4 paragraph 2) PNG media_image4.png 87 549 media_image4.png Greyscale It would have been obvious to a person having ordinary skill in the art before the effective filing date of the present invention to combine Tavazzi, Zhang, and Nikolova-Simons with Cini for the same reasons given in claim 7 above. Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Tavazzi in view of in view of Zhang, Nikolova-Simons, Cini, and further in view of NPL reference Chen et al. “Low-Rank Autoregressive Tensor Completion for Multivariate Time Series Forecasting” herein referred to as Chen. Regarding Claim 10 Tavazzi in view of Zhang, Nikolova-Simons, and Cini teaches: The method of claim 8 (see rejection of claim 8) Tavazzi in view of Zhang, Nikolova-Simons, and Cini does not teach: characterized in that, as an imputation model, Low-Rank Autoregressive Tensor Completion (LATC) is applied, which is constructed based on an autoregressive model that may consider the overall data flow of time series data, and converts the time series data, which is multivariate, into a three-dimensional tensor form and applies it to the autoregressive model. However, Chen teaches: characterized in that, as an imputation model, Low-Rank Autoregressive Tensor Completion (LATC) is applied, which is constructed based on an autoregressive model that may consider the overall data flow of time series data, and converts the time series data, which is multivariate, into a three-dimensional tensor form and applies it to the autoregressive model. (page 1 abstract) “The key of LATC is to transform the original multivariate time series matrix (e.g., sensor×time point) to a third-order tensor structure (e.g., sensor×time of day×day) by introducing an additional temporal dimension, which allows us to model the inherent rhythms and seasonality of time series as global patterns. With the tensor structure, we can transform the time series prediction and missing data imputation problems into a universal low-rank tensor completion problem.” Tavazzi, Zhang Nikolova-Simons, Cini, Chen, and the instant application are analogous because they are all directed to data processing. It would have been obvious to a person having ordinary skill in the art before the effective filing date of the present invention to modify the network communication log and imputations of Tavazzi in view of Zhang, Nikolova-Simons, and Cini by applying the LATC model of Chen for imputation because (Chen page 1 abstract) “Our numerical experiments on three real-world data sets demonstrate the superiority of the integration of global and local trends in LATC in both missing data imputation and rolling prediction tasks.” Claim 11 is rejected under 35 U.S.C. 103 as being unpatentable over Tavazzi in view of Zhang, Nikolova-Simons, and further in view of Chang et al. (US 20200081759 A1) herein referred to as Chang. Regarding Claim 11 Tavazzi in view of Zhang, Nikolova-Simons teaches: with imputed missing values by the method of claim 6 (see rejection of claim 6) Tavazzi in view of Zhang, Nikolova-Simons does not explicitly teach: A method for classifying root cause of defect by processing missing value in network log data, characterized in that a Fully Connected Neural Network (FCNN) is trained by non-time series data […] to obtain a model with optimal parameters, which is then used to classify the root cause of the defect for the non-time series data. However, Chang teaches: A method for classifying root cause of defect by processing missing value in network log data, characterized in that a Fully Connected Neural Network (FCNN) is trained by non-time series data […] to obtain a model with optimal parameters, which is then used to classify the root cause of the defect for the non-time series data. (paragraph [0111]) “FIG. 10 illustrates a flowchart of an example method for detecting a root-cause of the defect identified in the input data, according to one or more embodiments of the present invention. The data correction apparatus 120 compares the identified defect in the input data with root-cause classification models, as shown in 1020. In one or more examples, the comparison is performed using artificial neural networks, or other machine learning techniques.” Tavazzi, Zhang, Nikola-Simons, Chang, and the instant application are analogous because they are all directed to data processing. It would have been obvious to a person having ordinary skill in the art before the effective filing date of the present invention to modify the network communication log and imputations of Kim in view of Zhang and Cini by using it for the FCNN and root-cause analysis as taught by Chang because (Chang paragraph [0123]) “Further yet, the system includes a root-cause analysis module/subsection that uses machine learning classification algorithms (kNN, clustering, neural networks, etc.) to determine which errors exist in the input data and predictions and to classify the likelihood of these errors.” Further, advanced data imputation provides the benefit of making use of all available information. Claim 12 is rejected under 35 U.S.C. 103 as being unpatentable over Tavazzi in view of Zhang, Nikolova-Simons, Cini, and further in view of NPL reference Oğuz et al. “Anomaly detection in multi-tiered cellular networks using LSTM and 1D CNN” herein referred to as Oğuz. Regarding Claim 12 Tavazzi in view of Zhang, Nikolova-Simons, and Cini teaches: by the method of claim 7 (see rejection of claim 7) Tavazzi in view of Zhang, Nikolova-Simons, and Cini does not explicitly teach: A method for classifying root causes of defects by processing missing value in network log data, characterized in that a 1-Dimensional Convolutional Neural Network (1D-CNN) is trained by time series data with imputed missing values […] to obtain a model with optimal parameters, which is then used to classify the root cause of the defect for the time series data. However, Oğuz teaches: A method for classifying root causes of defects by processing missing value in network log data, characterized in that a 1-Dimensional Convolutional Neural Network (1D-CNN) is trained by time series data with imputed missing values […] to obtain a model with optimal parameters, which is then used to classify the root cause of the defect for the time series data. (page 3 second to last paragraph) “This study aims to develop a foresight for cell outage management in existing cellular systems and forthcoming next-generation networks within multi-tiered ultra-dense deployments. Inspired by the advancements of deep learning methods on time sequence analysis, we propose using long short-term memory and one-dimensional CNN, concomitantly with aggregation decision methods, for detection of femtocell outages which pose extra challenges. In the proposed scheme, we suggest not only detecting femtocell outages but also adroitly classifying the type of anomalies into three outage subclasses according to the severity of the degradation of the femto access point functionality.” Tavazzi, Zhang, Nikolova-Simons, Cini. Oğuz, and the instant application are analogous because they are all directed to data processing. It would have been obvious to a person having ordinary skill in the art before the effective filing date of the present invention to modify the network communication log and imputations of Tavazzi in view of Zhang , Nikolova-Simons and Cini by using it for the root cause analysis as taught by Oğuz because (Oğuz page 1 abstract) “Our findings show that proposed deep learning approaches outperform existing feed-forward neural networks, and on the average, in more than 80% of the cases the outage states of the femtocells are correctly predicted among healthy and three anomalous states.” Conclusion Any inquiry concerning this communication or earlier communications from the examiner should be directed to Ezra J Baker whose telephone number is (703)756-1087. The examiner can normally be reached Monday - Friday 10:00 am - 8:00 pm ET. 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, David Yi can be reached at (571) 270-7519. 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. /E.J.B./Examiner, Art Unit 2126 /DAVID YI/Supervisory Patent Examiner, Art Unit 2126
Read full office action

Prosecution Timeline

Nov 07, 2023
Application Filed
May 22, 2026
Non-Final Rejection mailed — §103, §112
Aug 24, 2026
Response Filed
Oct 30, 2026
Non-Final Rejection mailed — §103, §112 (current)

Precedent Cases

Applications granted by this same examiner with similar technology

Patent 12769999
APPARATUS AND METHOD FOR NEURAL NETWORK LEARNING USING SYNAPSE BASED ON MULTI ELEMENT
4y 10m to grant Granted Oct 20, 2026
Patent 12725073
METADATA GENERATION FOR INTELLIGENT ORCHESTRATION
3y 2m to grant Granted Sep 01, 2026
Patent 12718075
METHODS, SYSTEMS, AND MEDIA FOR COMPUTER VISION USING 2D CONVOLUTION OF 4D VIDEO DATA TENSORS
4y 10m to grant Granted Aug 25, 2026
Patent 12675712
Identity Graphing for Network Genomes
5y 0m to grant Granted Jul 07, 2026
Patent 12619886
Frozen Model Adaptation Through Soft Prompt Transfer
3y 9m to grant Granted May 05, 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

2-3
Expected OA Rounds
47%
Grant Probability
66%
With Interview (+19.6%)
4y 2m (~1y 2m remaining)
Median Time to Grant
Moderate
PTA Risk
Based on 30 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