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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 11/12/2024 and 10/20/2025 was filed. The submission is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Status of the Claims
Claims 1-19 have been examined.
Claim Rejections - 35 USC § 112
Claims 1-16 is/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. Claim 1 and 9 discloses “…a previous time series sketch of the sequence of time series sketches…”, it’s unclear, because the first batch lacks a previous sketch to reference, the claim fails to explain how initial weights are assigned or whether happens before, during or after a sketch is created. Appropriate correction required.
Claim 11 is/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. Claim 11 discloses “…the sequence of TCN-generated…”, but claim 11 depends on claim 9 which does not define/discloses the “TCN”, while claim 10 first time does define/discloses “TCN”. Therefore claim 11 lacks the antecedent basis for “…the sequence of TCN-generated…”. Appropriate correction required.
Claims 4-5 and 12-13 is/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 4 and 12 discloses “…generating a sequence of encoded time batches…” while claims 5 and 13 discloses “…the sequence of encoded time series batches…”. Therefore claims 5 and 13 lacks the antecedent basis for “…generating a sequence of encoded time batches…”. Appropriate correction required.
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 1-19 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
101 Analysis - Step 1
Claims 1-8 is/are recite a method/process, therefore claims 1-8 are within at least one of the four statutory categories.
Claims 9-16 and 17-19 is/are recite an apparatus/machine, therefore claims 9-16 and 17-19 are within at least one of the four statutory categories.
101 Analysis - Step 2A, Prong 1
Regarding Prong 1 of the Step 2A analysis in the 2019 PEG, the claims are to be analyzed to determine whether they recite subject matter that falls within one of the follow groups of abstract ideas: a) mathematical concepts, b) certain methods of organizing human activity, and/or c) mental processes.
Independent claim 11 includes limitations that recites mental processes and/or mathematical concepts (emphasized below) and will be used as a representative claim for the remainder of the 101 rejection. Claim 11 recites:
A computing system for analyzing time series data, the system comprising:
one or more processors; and one or more non-transitory computer-readable media that collectively store instructions that, when executed by the one or more processors, cause the computing system to perform operations, the operations comprising:
obtaining a sequence of time series batches of the time series data;
generating, by an Attentive Power Iteration (API) model, a sequence of time series sketches based on the sequence of time series batches of the time series data;
assigning a weight to new time series batches in the sequence of time series batches based at least partially on a previous time series sketch of the sequence of time series sketches; and
generating an output for each time series sketch in the sequence of time series sketches.
These limitations, as drafted, is a system that, under its broadest reasonable interpretation, covers performance of the limitation as a mental process and/or mathematical concept. That is, nothing in the claim elements preclude the steps from practically being performed as mathematical concepts. For example, " obtaining a sequence of time series …" and " generating, by an Attentive Power Iteration (API) model...", and “assigning a weight to new time series batches in the sequence …” and “generating an output for each time series…” discloses mathematical calculation and/or a mental process nature data analysis and evaluation. Therefore, these falls directly within the Mathematical Concepts and Mental processes grouping of abstract ideas.
101 Analysis - Step 2A, Prong 2
Regarding Prong 2 of the Step 2A analysis in the 2019 PEG, the claims are to be analyzed to determine whether the claim, as a whole, integrates the abstract idea into a practical application. As noted in the 2019 PEG, it must be determined whether any additional elements in the claim beyond the abstract idea integrate the exception into a practical application in a manner that imposes a meaningful limit on the judicial exception. The courts have indicated that additional elements merely using a computer to implement an abstract idea, adding insignificant extra solution activity, or generally linking use of a judicial exception to a particular technological environment or field of use do not integrate a judicial exception into a "practical application."
In the present case, the additional limitations beyond the above-noted abstract idea are as follows (where the underlined portions are the "additional limitations" while the bolded portions continue to represent the "abstract idea"):
A computing system for analyzing time series data, the system comprising:
one or more processors; and one or more non-transitory computer-readable media that collectively store instructions that, when executed by the one or more processors, cause the computing system to perform operations, the operations comprising:
obtaining a sequence of time series batches of the time series data;
generating, by an Attentive Power Iteration (API) model, a sequence of time series sketches based on the sequence of time series batches of the time series data;
assigning a weight to new time series batches in the sequence of time series batches based at least partially on a previous time series sketch of the sequence of time series sketches; and
generating an output for each time series sketch in the sequence of time series sketches.
For the following reason(s), the examiner submits that the above identified additional limitations do not integrate the above-noted abstract idea into a practical application.
Regarding the additional limitations of " one or more processors …" and “one or more non-transitory computer-readable media …” the components are merely generic components to perform a function using computer code. The generic components are recited at a high level of generality (i.e. a generic processor and memory) such that it amounts to no more than mere instructions to apply the exception using generic computer components. The examiner submits that these limitations are merely applying the above-noted abstract idea by merely using a general controller to perform the process (MPEP §2106.05).
Thus, taken alone, the additional elements do not integrate the abstract idea into a practical application. Further, looking at the additional limitation(s) as an ordered combination or as a whole, the limitation(s) add nothing that is not already present when looking at the elements taken individually. For instance, there is no indication that the additional elements, when considered as a whole, reflect an improvement in the functioning or an improvement to another technology or technical field, apply or use the above-noted judicial exception to effect a particular process for safety performance evaluation, implement/use the above-noted judicial exception with a particular machine or manufacture that is integral to the claim, effect a transformation or reduction of a particular article to a different state or thing, or apply or use the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment, such that the claim as a whole is not more than a drafting effort designed to monopolize the exception (MPEP § 2106.05). Accordingly, the additional limitation(s) do/does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
101 Analysis - Step 2B
Regarding Step 2B in the 2019 PEG, representative independent claim 12 does not include additional elements (considered both individually and as an ordered combination) that are sufficient to amount to significantly more than the judicial exception for the same reasons to those discussed above with respect to determining that the claim does not integrate the abstract idea into a practical application. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of " one or more processors …" and “one or more non-transitory computer-readable media …” amounts to nothing more than applying the exception using a generic computer component. Mere instructions cannot provide an inventive concept. Hence, the claim is not patent eligible.
Claims 1 and 17 recites analogous limitations to that of claim 9, and are therefore rejected by the same premise.
Dependent claims 2-8, 10-16, and 18-19 specify limitations that elaborate on the abstract idea of claims 1, 9 and 17, and thus are directed to an abstract idea nor do the claims recite additional limitations that integrate the claims into a practical application or amount to “significantly more” for similar reasons.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1-19 is/are rejected under 35 U.S.C. 102(a)(1) as being unpatentable over Hao (Deep Time Series Sketching and Its Application on Industrial Time Series Clustering XP034283075 • 2022-12-17 • Huang Hao + 2).
Claim.1 Hao discloses a computer-implemented method for analyzing time series data (see at least abstract, voluminous multivariate time series data collected from sensors provides tremendous benefit for understanding of modern industrial systems such as power plants, wind turbines and aircrafts, time series clustering has become one of the key analysis techniques), the method comprising: obtaining, by a computing system comprising one or more processors a sequence of time series batches of the time series data (see at least fig.1-3, abstract, a representation learning model based on temporal convolutional networks that perform on a sliding window basis along time series to learn the windows’ embeddings, the sequence of embeddings is then fed into embedding sketching to obtain its sketch, page 1997, large volume of time series data are collected by various sensors in industrial systems such as power plants, aircrafts and wind turbines. These sensors measure different variables of the systems in high frequency over time, resulting in a vast mass of data mostly impractical for a manual analysis); generating, by the computing system and with an Attentive Power Iteration (API) model, a sequence of time series sketches based on the sequence of time series batches of the time series data (see at least fig.1-6, abstract, Deep Time Series Sketching (DTSS) model. This model is a representation learning model based on temporal convolutional networks that perform on a sliding window basis along time series to learn the windows’ embeddings. The sequence of embeddings is then fed into embedding sketching to obtain its sketch. Such sketch is a descriptor of the whole time series and will be fed into K-means for clustering, page 1997, large volume of time series data are collected by various sensors in industrial systems such as power plants, aircrafts and wind turbines. These sensors measure different variables of the systems in high frequency over time, resulting in a vast mass of data mostly impractical for a manual analysis, conventional time series clustering focus on learning time series representation or similarity metric in raw variable space); assigning, by the computing system, a weight to new time series batches in the sequence of time series batches based at least partially on a previous time series sketch of the sequence of time series sketches (see at least fig.1-6, page 1998, a deep learning and embedding sketching based model, Deep Time Series Sketching (DTSS), for clustering industrial multivariate time series with varying lengths. It is designed as a representation learning method that takes input from time series in a sliding window manner. Nonlinear features of each window is extracted and used to update the time series sketch as the window slides from the beginning to the end of the time series. As shown in Figure 2, our DTSS consists of two parts: temporal convolutional networks and embedding sketching. While the encoding form of temporal convolutional networks pays more attention to the local characteristics in each window, the embedding sketching part considers and synthesizes the complete set of patterns along all windows, and outputs a balanced and concise representation that contains both global and salient local characteristics, page 1999, the sequence of embeddings obtained from TCN bottle neck layer is fed into embedding sketching to learn the sketch of embeddings. The sketch captures the important structure by considering both frequently appearing embeddings and certain salient embeddings regardless of time series length); and generating, by the computing system, an output for each time series sketch in the sequence of time series sketches (see at least fig.1-6, page 1998, the embedding sketching part considers and synthesizes the complete set of patterns along all windows, and outputs a balanced and concise representation that contains both global and salient local characteristics. Furthermore, our DTSS can be used in a real time streaming, meaning that, unlike the other algorithms, DTSS doesn’t require to see the whole time series to assign labels. Instead, it is capable to output the up-to-date representation and label of the time series in real time, page 1999, embedding sketching is an incrementally update process, it can output up-to-date sketch for a time series as long as the model has seen a small number of windows. Therefore we can run K-means on all the currently available sketches and obtain their labels in real time).
Claim.2 Hao discloses further comprising providing the sequence of time series batches to a Temporal Convolutional Network (TCN) to generate a sequence of TCN-generated embeddings, wherein the TCN extracts nonlinear features from the sequence of time series batches by performing sliding window processing on the time series data (see at least fig.1-6, abstract, Deep Time Series Sketching (DTSS) model. This model is a representation learning model based on temporal convolutional networks that perform on a sliding window basis along time series to learn the windows’ embeddings. The sequence of embeddings is then fed into embedding sketching to obtain its sketch. Such sketch is a descriptor of the whole time series and will be fed into K-means for clustering, page 1997, large volume of time series data are collected by various sensors in industrial systems such as power plants, aircrafts and wind turbines. These sensors measure different variables of the systems in high frequency over time, resulting in a vast mass of data mostly impractical for a manual analysis, conventional time series clustering focus on learning time series representation or similarity metric in raw variable space, page 1999, the sequence of embeddings obtained from TCN bottle neck layer is fed into embedding sketching to learn the sketch of embeddings. The sketch captures the important structure by considering both frequently appearing embeddings and certain salient embeddings regardless of time series length).
Claim.3 Hao discloses further comprising providing the sequence of TCN-generated embeddings to the API model as an input (see at least fig.1-6, abstract, Deep Time Series Sketching (DTSS) model. This model is a representation learning model based on temporal convolutional networks that perform on a sliding window basis along time series to learn the windows’ embeddings. The sequence of embeddings is then fed into embedding sketching to obtain its sketch. Such sketch is a descriptor of the whole time series and will be fed into K-means for clustering, page 1997, large volume of time series data are collected by various sensors in industrial systems such as power plants, aircrafts and wind turbines. These sensors measure different variables of the systems in high frequency over time, resulting in a vast mass of data mostly impractical for a manual analysis, conventional time series clustering focus on learning time series representation or similarity metric in raw variable space, page 1999, the sequence of embeddings obtained from TCN bottle neck layer is fed into embedding sketching to learn the sketch of embeddings. The sketch captures the important structure by considering both frequently appearing embeddings and certain salient embeddings regardless of time series length).
Claim.4 Hao discloses further comprising providing the sequence of time series batches to a positional encoding model of the API model, the positional encoding model generating a sequence of encoded time batches from the sequence of time series batches (see at least fig.1-6, abstract, Deep Time Series Sketching (DTSS) model. This model is a representation learning model based on temporal convolutional networks that perform on a sliding window basis along time series to learn the windows’ embeddings. The sequence of embeddings is then fed into embedding sketching to obtain its sketch. Such sketch is a descriptor of the whole time series and will be fed into K-means for clustering, page 1997, large volume of time series data are collected by various sensors in industrial systems such as power plants, aircrafts and wind turbines. These sensors measure different variables of the systems in high frequency over time, resulting in a vast mass of data mostly impractical for a manual analysis, conventional time series clustering focus on learning time series representation or similarity metric in raw variable space).
Claim.5 Hao discloses further comprising providing the sequence of encoded time series batches to a projection model, the projection model generating the sequence of time series sketches from the sequence of encoded time series batches (see at least fig.1-6, abstract, Deep Time Series Sketching (DTSS) model. This model is a representation learning model based on temporal convolutional networks that perform on a sliding window basis along time series to learn the windows’ embeddings. The sequence of embeddings is then fed into embedding sketching to obtain its sketch. Such sketch is a descriptor of the whole time series and will be fed into K-means for clustering, page 1997, large volume of time series data are collected by various sensors in industrial systems such as power plants, aircrafts and wind turbines. These sensors measure different variables of the systems in high frequency over time, resulting in a vast mass of data mostly impractical for a manual analysis, conventional time series clustering focus on learning time series representation or similarity metric in raw variable space).
Claim.6 Hao discloses further comprising providing the sequence of time series sketches to a classification block to predict a class probability distribution (see at least fig.1-6, page 1998, these methods are either linear extraction that cannot capture nonlinear information, or based on certain distribution assumption, deep learning based methods are gaining more and more popularity due to its capability to perform automatic high-level feature extraction without explicit assumption on data distribution, and its high prediction accuracy when trained with huge amount of data. Recurrent Neural Network (RNN) is one of the most widely used deep learning techniques in time series learning).
Claim.7 Hao discloses wherein generating an output further comprises: generating a classification label for each time series sketch in the sequence of time series sketches (see at least fig.1-6, page 1998, a deep learning and embedding sketching based model, Deep Time Series Sketching (DTSS), for clustering industrial multivariate time series with varying lengths. It is designed as a representation learning method that takes input from time series in a sliding window manner. Nonlinear features of each window is extracted and used to update the time series sketch as the window slides from the beginning to the end of the time series. As shown in Figure 2, our DTSS consists of two parts: temporal convolutional networks and embedding sketching. While the encoding form of temporal convolutional networks pays more attention to the local characteristics in each window, the embedding sketching part considers and synthesizes the complete set of patterns along all windows, and outputs a balanced and concise representation that contains both global and salient local characteristics, page 1999, the sequence of embeddings obtained from TCN bottle neck layer is fed into embedding sketching to learn the sketch of embeddings. The sketch captures the important structure by considering both frequently appearing embeddings and certain salient embeddings regardless of time series length).
Claim.8 Hao discloses wherein generating an output further comprises: generating a prediction for each time series sketch in the sequence of time series sketches (see at least fig.1-6, page 1998, a deep learning and embedding sketching based model, Deep Time Series Sketching (DTSS), for clustering industrial multivariate time series with varying lengths. It is designed as a representation learning method that takes input from time series in a sliding window manner. Nonlinear features of each window is extracted and used to update the time series sketch as the window slides from the beginning to the end of the time series. As shown in Figure 2, our DTSS consists of two parts: temporal convolutional networks and embedding sketching. While the encoding form of temporal convolutional networks pays more attention to the local characteristics in each window, the embedding sketching part considers and synthesizes the complete set of patterns along all windows, and outputs a balanced and concise representation that contains both global and salient local characteristics, page 1999, the sequence of embeddings obtained from TCN bottle neck layer is fed into embedding sketching to learn the sketch of embeddings. The sketch captures the important structure by considering both frequently appearing embeddings and certain salient embeddings regardless of time series length).
Claim.9 Hao discloses a computing system for analyzing time series data (see at least abstract, voluminous multivariate time series data collected from sensors provides tremendous benefit for understanding of modern industrial systems such as power plants, wind turbines and aircrafts, time series clustering has become one of the key analysis techniques), the system comprising: one or more processors; and one or more non-transitory computer-readable media that collectively store instructions that, when executed by the one or more processors, cause the computing system to perform operations, the operation comprising: obtaining a sequence of time series batches of time series data (see at least fig.1-3, abstract, a representation learning model based on temporal convolutional networks that perform on a sliding window basis along time series to learn the windows’ embeddings, the sequence of embeddings is then fed into embedding sketching to obtain its sketch, page 1997, large volume of time series data are collected by various sensors in industrial systems such as power plants, aircrafts and wind turbines. These sensors measure different variables of the systems in high frequency over time, resulting in a vast mass of data mostly impractical for a manual analysis); generating, by an Attentive Power Iteration (API) model, a sequence of time series sketches based on the sequence of time series batches of time series data (see at least fig.1-6, abstract, Deep Time Series Sketching (DTSS) model. This model is a representation learning model based on temporal convolutional networks that perform on a sliding window basis along time series to learn the windows’ embeddings. The sequence of embeddings is then fed into embedding sketching to obtain its sketch. Such sketch is a descriptor of the whole time series and will be fed into K-means for clustering, page 1997, large volume of time series data are collected by various sensors in industrial systems such as power plants, aircrafts and wind turbines. These sensors measure different variables of the systems in high frequency over time, resulting in a vast mass of data mostly impractical for a manual analysis, conventional time series clustering focus on learning time series representation or similarity metric in raw variable space); assigning a weight to new time series batches in the sequence of time series batches based at least partially on a previous time series sketch of the sequence of time series sketches (see at least fig.1-6, page 1998, a deep learning and embedding sketching based model, Deep Time Series Sketching (DTSS), for clustering industrial multivariate time series with varying lengths. It is designed as a representation learning method that takes input from time series in a sliding window manner. Nonlinear features of each window is extracted and used to update the time series sketch as the window slides from the beginning to the end of the time series. As shown in Figure 2, our DTSS consists of two parts: temporal convolutional networks and embedding sketching. While the encoding form of temporal convolutional networks pays more attention to the local characteristics in each window, the embedding sketching part considers and synthesizes the complete set of patterns along all windows, and outputs a balanced and concise representation that contains both global and salient local characteristics, page 1999, the sequence of embeddings obtained from TCN bottle neck layer is fed into embedding sketching to learn the sketch of embeddings. The sketch captures the important structure by considering both frequently appearing embeddings and certain salient embeddings regardless of time series length); and generating an output for each time series sketch in the sequence of time series sketches (see at least fig.1-6, page 1998, the embedding sketching part considers and synthesizes the complete set of patterns along all windows, and outputs a balanced and concise representation that contains both global and salient local characteristics. Furthermore, our DTSS can be used in a real time streaming, meaning that, unlike the other algorithms, DTSS doesn’t require to see the whole time series to assign labels. Instead, it is capable to output the up-to-date representation and label of the time series in real time, page 1999, embedding sketching is an incrementally update process, it can output up-to-date sketch for a time series as long as the model has seen a small number of windows. Therefore we can run K-means on all the currently available sketches and obtain their labels in real time).
Claim.10 Hao discloses further comprising providing the sequence of time series batches to a Temporal Convolutional Network (TCN) to generate a sequence of TCN-generated embeddings, wherein the TCN employs a sliding window to extract nonlinear features from the sequence of time series batches (see at least fig.1-6, abstract, Deep Time Series Sketching (DTSS) model. This model is a representation learning model based on temporal convolutional networks that perform on a sliding window basis along time series to learn the windows’ embeddings. The sequence of embeddings is then fed into embedding sketching to obtain its sketch. Such sketch is a descriptor of the whole time series and will be fed into K-means for clustering, page 1997, large volume of time series data are collected by various sensors in industrial systems such as power plants, aircrafts and wind turbines. These sensors measure different variables of the systems in high frequency over time, resulting in a vast mass of data mostly impractical for a manual analysis, conventional time series clustering focus on learning time series representation or similarity metric in raw variable space, page 1999, the sequence of embeddings obtained from TCN bottle neck layer is fed into embedding sketching to learn the sketch of embeddings. The sketch captures the important structure by considering both frequently appearing embeddings and certain salient embeddings regardless of time series length).
Claim.11 Hao discloses further comprising providing the sequence of TCN-generated embeddings to the API model as an input (see at least fig.1-6, abstract, Deep Time Series Sketching (DTSS) model. This model is a representation learning model based on temporal convolutional networks that perform on a sliding window basis along time series to learn the windows’ embeddings. The sequence of embeddings is then fed into embedding sketching to obtain its sketch. Such sketch is a descriptor of the whole time series and will be fed into K-means for clustering, page 1997, large volume of time series data are collected by various sensors in industrial systems such as power plants, aircrafts and wind turbines. These sensors measure different variables of the systems in high frequency over time, resulting in a vast mass of data mostly impractical for a manual analysis, conventional time series clustering focus on learning time series representation or similarity metric in raw variable space, page 1999, the sequence of embeddings obtained from TCN bottle neck layer is fed into embedding sketching to learn the sketch of embeddings. The sketch captures the important structure by considering both frequently appearing embeddings and certain salient embeddings regardless of time series length).
Claim.12 Hao discloses further comprising providing the sequence of time series batches to a positional encoding model of the API, the positional encoding model generating a sequence of encoded time batches from the sequence of time series batches (see at least fig.1-6, abstract, Deep Time Series Sketching (DTSS) model. This model is a representation learning model based on temporal convolutional networks that perform on a sliding window basis along time series to learn the windows’ embeddings. The sequence of embeddings is then fed into embedding sketching to obtain its sketch. Such sketch is a descriptor of the whole time series and will be fed into K-means for clustering, page 1997, large volume of time series data are collected by various sensors in industrial systems such as power plants, aircrafts and wind turbines. These sensors measure different variables of the systems in high frequency over time, resulting in a vast mass of data mostly impractical for a manual analysis, conventional time series clustering focus on learning time series representation or similarity metric in raw variable space).
Claim.13 Hao discloses further comprising providing the sequence of encoded time series batches to a projection model, the projection model generating the sequence of time series sketches from the sequence of encoded time series batches (see at least fig.1-6, abstract, Deep Time Series Sketching (DTSS) model. This model is a representation learning model based on temporal convolutional networks that perform on a sliding window basis along time series to learn the windows’ embeddings. The sequence of embeddings is then fed into embedding sketching to obtain its sketch. Such sketch is a descriptor of the whole time series and will be fed into K-means for clustering, page 1997, large volume of time series data are collected by various sensors in industrial systems such as power plants, aircrafts and wind turbines. These sensors measure different variables of the systems in high frequency over time, resulting in a vast mass of data mostly impractical for a manual analysis, conventional time series clustering focus on learning time series representation or similarity metric in raw variable space).
Claim.14 Hao discloses further comprising providing the sequence of time series sketches to a classification block to predict a class probability distribution, the classification block comprising a fully connected layer and a softmax function (see at least fig.1-6, page 1998, these methods are either linear extraction that cannot capture nonlinear information, or based on certain distribution assumption, deep learning based methods are gaining more and more popularity due to its capability to perform automatic high-level feature extraction without explicit assumption on data distribution, and its high prediction accuracy when trained with huge amount of data. Recurrent Neural Network (RNN) is one of the most widely used deep learning techniques in time series learning).
Claim.15 Hao discloses wherein generating an output further comprises: generating a classification label for each time series sketch in the sequence of time series sketches (see at least fig.1-6, page 1998, a deep learning and embedding sketching based model, Deep Time Series Sketching (DTSS), for clustering industrial multivariate time series with varying lengths. It is designed as a representation learning method that takes input from time series in a sliding window manner. Nonlinear features of each window is extracted and used to update the time series sketch as the window slides from the beginning to the end of the time series. As shown in Figure 2, our DTSS consists of two parts: temporal convolutional networks and embedding sketching. While the encoding form of temporal convolutional networks pays more attention to the local characteristics in each window, the embedding sketching part considers and synthesizes the complete set of patterns along all windows, and outputs a balanced and concise representation that contains both global and salient local characteristics, page 1999, the sequence of embeddings obtained from TCN bottle neck layer is fed into embedding sketching to learn the sketch of embeddings. The sketch captures the important structure by considering both frequently appearing embeddings and certain salient embeddings regardless of time series length).
Claim.16 Hao discloses wherein generating an output further comprises: generating a prediction for each time series sketch in the sequence of time series sketches (see at least fig.1-6, page 1998, a deep learning and embedding sketching based model, Deep Time Series Sketching (DTSS), for clustering industrial multivariate time series with varying lengths. It is designed as a representation learning method that takes input from time series in a sliding window manner. Nonlinear features of each window is extracted and used to update the time series sketch as the window slides from the beginning to the end of the time series. As shown in Figure 2, our DTSS consists of two parts: temporal convolutional networks and embedding sketching. While the encoding form of temporal convolutional networks pays more attention to the local characteristics in each window, the embedding sketching part considers and synthesizes the complete set of patterns along all windows, and outputs a balanced and concise representation that contains both global and salient local characteristics, page 1999, the sequence of embeddings obtained from TCN bottle neck layer is fed into embedding sketching to learn the sketch of embeddings. The sketch captures the important structure by considering both frequently appearing embeddings and certain salient embeddings regardless of time series length).
Claim.17 Hao discloses a computing system for time-based data classification (see at least abstract, voluminous multivariate time series data collected from sensors provides tremendous benefit for understanding of modern industrial systems such as power plants, wind turbines and aircrafts, time series clustering has become one of the key analysis techniques), the system comprising: one or more processors; and one or more non-transitory computer-readable media that collectively store instructions that, when executed by the one or more processors, cause the computing system to perform operations, the operations comprising: obtaining time series data, wherein the time series data is descriptive of a series of time-based data points (see at least fig.1-3, abstract, a representation learning model based on temporal convolutional networks that perform on a sliding window basis along time series to learn the windows’ embeddings, the sequence of embeddings is then fed into embedding sketching to obtain its sketch, page 1997, large volume of time series data are collected by various sensors in industrial systems such as power plants, aircrafts and wind turbines. These sensors measure different variables of the systems in high frequency over time, resulting in a vast mass of data mostly impractical for a manual analysis); processing the time series data with a temporal convolution model to generate one or more time series embeddings (see at least fig.1-6, abstract, Deep Time Series Sketching (DTSS) model. This model is a representation learning model based on temporal convolutional networks that perform on a sliding window basis along time series to learn the windows’ embeddings. The sequence of embeddings is then fed into embedding sketching to obtain its sketch. Such sketch is a descriptor of the whole time series and will be fed into K-means for clustering, page 1997, large volume of time series data are collected by various sensors in industrial systems such as power plants, aircrafts and wind turbines. These sensors measure different variables of the systems in high frequency over time, resulting in a vast mass of data mostly impractical for a manual analysis, conventional time series clustering focus on learning time series representation or similarity metric in raw variable space); processing the one or more time series embeddings with a representation generation model to generate an output representation, wherein the output representation generation model to generate an output representation, wherein the output representation is descriptive of a graphical representation of the series of time-based data points (see at least fig.1-6, page 1998, a deep learning and embedding sketching based model, Deep Time Series Sketching (DTSS), for clustering industrial multivariate time series with varying lengths. It is designed as a representation learning method that takes input from time series in a sliding window manner. Nonlinear features of each window is extracted and used to update the time series sketch as the window slides from the beginning to the end of the time series. As shown in Figure 2, our DTSS consists of two parts: temporal convolutional networks and embedding sketching. While the encoding form of temporal convolutional networks pays more attention to the local characteristics in each window, the embedding sketching part considers and synthesizes the complete set of patterns along all windows, and outputs a balanced and concise representation that contains both global and salient local characteristics, page 1999, the sequence of embeddings obtained from TCN bottle neck layer is fed into embedding sketching to learn the sketch of embeddings. The sketch captures the important structure by considering both frequently appearing embeddings and certain salient embeddings regardless of time series length); and processing the output representation with a classification model to generate a classification label for the time series data (see at least fig.1-6, page 1998, the embedding sketching part considers and synthesizes the complete set of patterns along all windows, and outputs a balanced and concise representation that contains both global and salient local characteristics. Furthermore, our DTSS can be used in a real time streaming, meaning that, unlike the other algorithms, DTSS doesn’t require to see the whole time series to assign labels. Instead, it is capable to output the up-to-date representation and label of the time series in real time, page 1999, embedding sketching is an incrementally update process, it can output up-to-date sketch for a time series as long as the model has seen a small number of windows. Therefore we can run K-means on all the currently available sketches and obtain their labels in real time).
Claim.18 Hao discloses wherein the output representation comprises a matrix sketch (see at least fig.1-6, page 2000, a sketch of matrix is an approximation of the original matrix but much smaller).
Claim.19 Hao discloses wherein the time series data was generated with one or more sensors associated with a turbine, and wherein the classification label comprises an anomaly detection classification (see at least fig.1-3, abstract, a representation learning model based on temporal convolutional networks that perform on a sliding window basis along time series to learn the windows’ embeddings, the sequence of embeddings is then fed into embedding sketching to obtain its sketch, page 1997, large volume of time series data are collected by various sensors in industrial systems such as power plants, aircrafts and wind turbines. These sensors measure different variables of the systems in high frequency over time, resulting in a vast mass of data mostly impractical for a manual analysis).
Conclusion
Related References
The relevant art made of record and not relied upon is considered pertinent to applicant’s disclosure.
Srivastava (US20250209084A1) discloses generating time series of sketches and using the time series of sketches for approximate query processing. In accordance with some aspects, tabular data is accessed that has a number of columns. Responsive to identifying a first column as comprising numerical data, sketches are generated for the numerical data for each of a number of time steps, and the sketches for the numerical data are stored as a first time series of sketches. Responsive to identifying a second column as comprising categorical data, sketches are generated for the categorical data for each of the time steps, and the sketches for the categorical data are stored as a second time series of sketches. When a query is received, a response to the query is provided using sketches from the first time series of sketches and/or the second time series of sketches.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SHARDUL D PATEL whose telephone number is (571)270-7758. The examiner can normally be reached Monday-Friday 8am-5pm (IFP).
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, KITO ROBINSON can be reached at (571)270-3921. 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.
/SHARDUL D PATEL/Primary Examiner, Art Unit 3664
Prosecution Insights